# Stanford CS 224N | Project Reports

https://web.stanford.edu/class/cs224n/project.html

[CS224N Home](https://web.stanford.edu/class/cs224n/index.html)
  * [Coursework](https://web.stanford.edu/class/cs224n/index.html#coursework)
  * [Schedule](https://web.stanford.edu/class/cs224n/index.html#schedule)
  * [Office Hours](https://web.stanford.edu/class/cs224n/office_hours.html)
  * [Final projects](https://web.stanford.edu/class/cs224n/project.html)
  * [Lecture Videos](https://canvas.stanford.edu/courses/191439/external_tools/3367)
  * [Ed Forum](https://edstem.org/us/courses/57406)


[ ![](https://web.stanford.edu/class/cs224n/images/stanford-nlp-logo-new.jpg) ](http://nlp.stanford.edu/) [ ![](https://web.stanford.edu/class/cs224n/images/stanfordlogo.jpg) ](http://stanford.edu/)
# CS224N: Natural Language Processing with Deep Learning
### Stanford / Winter 2026
## Project Awards
Congratulations to the following teams, who produced exceptional, award-winning projects! 
### Best Default Project
  * **[Fairness-Aware Fine-Tuning of GPT-2 for Paraphrase Detection](https://drive.google.com/file/d/1FmXex1CNP9NMM6Sn84QNOVNCnwjborKU/view?usp=sharing)**. Deonna Owens. 


### Best Custom Project
  * **[SYMBRION: Symbol Context and Dream Ego Relations Across Lifelong Dream Series as a Tool for Psychoanalysis](https://drive.google.com/file/d/1hdBkEktCljkEFAoe9Z8WTKKj-F92sRuh/view?usp=sharing)**. Bobby Rohrkemper, Chia-Wei Cheng. 


### Outstanding Default Projects
  * **[Accelerating Attention for GPT-2 Using FLASHATTENTION, Longformer, and cosFORMER](https://drive.google.com/file/d/1Laef75yBJxiepchqc4aptUHOSx3xpaub/view?usp=drive_link)**. Diego Sierra, Thomas Sarda, Tom-Eliot Jullien. 
  * **[KV Caching and Speculative Decoding at GPT-2 Scale: Acceptance, Cost Ratio, and the Limits of Speedup](https://drive.google.com/file/d/1Q4kw19wsL-EUwThLZnE5GdOS-5DrnTkf/view?usp=drive_link)**. Amulya Parthasarathy. 
  * **[Accelerated DPO Fine-tuning GPT-2 with Constructed Data](https://drive.google.com/file/d/1Duz7yR3nVvj5lhV2JdRWMcVUX28HOUoa/view?usp=sharing)**. Jessie Ou, Weixin Yu. 


### Outstanding Custom Projects
  * **[TRACE: Tool-augmented Reasoning via Atomic Cheatsheet Editing](https://drive.google.com/file/d/1iG2nPk_-tznWJohCS2C6MpHBHWDbZeh5/view?usp=sharing)**. Kyleen Liao, Roshen Nair, Arnold Yang. 
  * **[Dynamic Token Merging for Efficient Subword Encoder-Decoder Transformers](https://drive.google.com/file/d/1BXkobcZ22FagAsVXKMVEKggmwmcPSKaU/view?usp=drive_link)**. Nathan Zhou, Chris Gu, Marco Andono Sie. 
  * **[Concept Training for Human-Aligned Language Models](https://drive.google.com/file/d/1Yxh7W-q4wvc99DgoxAilTM4bW12JZCwh/view?usp=drive_link)**. Christine Zhang. 


## Custom Projects  
| Project name  | Authors  |  
| --- | --- |  
| [20 Questions for Code: Improving Code Generation with Information-Theoretic Clarification](https://drive.google.com/file/d/1TKtMp75zbRm2wDfSrtu3EHdYLGWzH7il/view?usp=sharing)  | Alexandra Suriya Kim, Julia Xi, Ria Garg  |  
| [A Bigger Catch: Fine-Grained Curriculum Standards Alignment on the MathFish Benchmark](https://drive.google.com/file/d/1dhGvOCGOEQkPm3TucPox5ZrSPUnvkkQ6/view?usp=drive_link)  | Mayank Sharma, Teah Shi, Xinman Liu  |  
| [Adapting Language Models for Low-Resource GPU Kernel Programming](https://drive.google.com/file/d/1JrCOwRFd5JgqjdTTi5C4hKOoDo3Gi5EH/view?usp=sharing)  | Annmaria Antony, Laasya Konidala, Natalia Pahlavan  |  
| [Adaptive Test-Time Compute for Efficient Reasoning in Language Models](https://drive.google.com/file/d/1bXK60si4tDGC61KY4J4RflhPTPL91uHS/view?usp=drive_link)  | Ryan Tan  |  
| [Adaptive Test-Time Compute for Pedagogically Grounded Reasoning in LLMs](https://drive.google.com/file/d/1OebpFnIki7nS_n96g95EpigCz5tdFIFQ/view?usp=drive_link)  | Isha Jain, Jason Sejin Chon, Medhya Goel  |  
| [Agents Don’t Always Do What They Think](https://drive.google.com/file/d/1ChlbaaCJUpQTPKm1oJYp1zxk-f_N7lbx/view?usp=drive_link)  | Mark William Gernitis  |  
| [Always-On Learning Companion: Proactive Multimodal Tutoring for Everyday Study Scenarios](https://drive.google.com/file/d/1DhGWTi_8jAl64mS5yBzQinCL_AOqCpfq/view?usp=drive_link)  | Chenyue Li, Haowen Wang, Zhen Jia  |  
| [Analyzing Robustness and Context Use in Clinical Natural Language Inference](https://drive.google.com/file/d/1H_awmB8Nxq4lgZiuimVqPDAdx-u_y4e2/view?usp=drive_link)  | Ryan Minh-Tri Le  |  
| [Attention Modifications for Improved Adaptation](https://drive.google.com/file/d/1zWrlZasrX3astiulkP3bbGlPKGO8u_TD/view?usp=sharing)  | Jerry Yin, Michael Jang  |  
| [Auditing Model-Generated Privacy Benchmarks: Do Synthetic Evaluations Reflect Real User Privacy Norms?](https://drive.google.com/file/d/1_s5o22mQEnUtyeEtSsG36ePE_pw4uy7O/view?usp=drive_link)  | Selena She  |  
| [Benchmarking and Improving Generative Diversity in Language Models via Diverse Preference Optimization](https://drive.google.com/file/d/1HFoqonVdYp49ocO0uSFkNQlKD-t8MFxB/view?usp=drive_link)  | Annika Kaul Singh, Shyam Sai Bethina  |  
| [Betting on Reasoning: Predicting Forecast Reliability in Prediction Markets](https://drive.google.com/file/d/1LTxZYsm_dlroDcriFxliZDcdW-xAN0MY/view)  | Rahul Rejeev  |  
| [Beyond Bradley-Terry: Random Logit Preference Modeling for RLHF](https://drive.google.com/file/d/1ox2zajKZTO6isZNTgXxZSXgerKweSv4r/view?usp=drive_link)  | Junyi Liu  |  
| [Beyond Knowledge: Syntactic Complexity as a Bottleneck for Reasoning in "Bracket City" Puzzles](https://drive.google.com/file/d/1z4zxXl7MJQyAAlrt1qfFdQIGxGS8j2zq/view?usp=drive_link)  | Amrita Malhotra  |  
| [Bootstrapping Reasoning in Compact Language Models: A Multi-Stage Reinforcement Learning Pipeline with Targeted Failure Repair](https://drive.google.com/file/d/1PbRVolm96aXGvKghIMs0CEYOEHZvLnK3/view?usp=sharing)  | Joseph Li, Max Luis Rodriguez, Victor Chen  |  
| [Bootstrapping Safety-Aligned Reasoning in Small Language Models via Self-Instruct](https://drive.google.com/file/d/1IVLPf80zT8-Ab_Sx_zk5EfgEcxqPCsqy/view?usp=sharing)  | J Yim, Komal Vij, Tim Jing  |  
| [Building A Contextual Reasoning Aware Social-Intel Agent with Reinforcement Learning](https://drive.google.com/file/d/1kcRAo9E_o2p9hsLf6M3paXsdCZRdq14J/view?usp=drive_link)  | Binbin Li, Da Sun, Ying Lu  |  
| [Burst: Multi-Agent System for High-Quality Temporal Content Generation](https://drive.google.com/file/d/1LvmLkgpG9_W50LOHWN4P5GEHXufZF8S2/view?usp=drive_link)  | Jeffrey Hao Wang  |  
| [Can Coding Agents Manage Their Own Memory?](https://drive.google.com/file/d/1-8wAcUiz3LpYMUavipmLfymO9p1PtCPM/view?usp=drive_link)  | Jerry Wang, Ryan Wang, Sameer Agrawal  |  
| [Cartoon Caption Humor Quality Assessment and Generation with DPO and LoRA](https://drive.google.com/file/d/1s8OGcFGctQXU-o5FL4gRUXjD1P8Caz2R/view?usp=sharing)  | Isaiah Flores, Katherine Ha Wang  |  
| [Causal Transfer of Semantic Operators Across Transformer Language Models](https://drive.google.com/file/d/1ojZBh9WmE7IPnMLGbW3_UiFiUZEPYlQ5/view)  | Shivatmica Murgai  |  
| [Childproofing LLMs with Contrastive Activation Addition](https://drive.google.com/file/d/1yy5HfOswguuKAFA3gzRtgqLtsdoyamqI/view)  | Rosemary Mingrui Jiang  |  
| [Childproofing LLMs: A Comparative Analysis of CAA, ReFT, and DPO for Safety Alignment](https://drive.google.com/file/d/1cQBOaYCWzgWWQdun8xVbBd-2Ss47e3_c/view)  | Alice Zhu, Anya Han Zhang  |  
| [Coaching Qwen3 Coder 30B to Think Like a CodeClash Arena Agent](https://drive.google.com/file/d/1pxtEYH6ZmY8EMD76IG8HeSTn-6fS1FuY/view?usp=sharing)  | Ivy (Ning) Zhang  |  
| [CoFi-PG: Counterfactual Policy Gradients under Filtered Feedback for Multi-Agent LLMs](https://drive.google.com/file/d/1yhgmpaip5_8iP-8cq6LUgI5ZYSUo3Owf/view?usp=sharing)  | Elai Ben-Gal, Stela Tong  |  
| [Cognitive Compression: Hierarchical Chain-of-Thought for Efficient LLM Reasoning](https://drive.google.com/file/d/1JaRHtanb6cGPZSin6NS1vmJBLdMkeWVH/view?usp=sharing)  | Anuj Jamwal  |  
| [Collaborative Dynamic Cheatsheet: Multi-Agent Test-Time Learning with Small Language Models](https://drive.google.com/file/d/1j0XWD31OgPDf137zQmR-3PTNh83YbNNM/view?usp=sharing)  | Erica Wang, Malvyn Lai  |  
| [Compiler-in-the-Loop: Decomposing the Value of Static Verification for Low-Resource Code Generation](https://drive.google.com/file/d/1TyfqWOj8RhC1z7FGMXWanvTW6WksGisG/view?usp=drive_link)  | Hlumelo Notshe, Joshua Martinez  |  
| [Compositional Tool-Sequence Generation in Small Language Models](https://drive.google.com/file/d/1e045Zn3wrk5y2boyIt3KkYryNsD-W2ZB/view?usp=drive_link)  | Benji Warburton, Maanit Goel  |  
| [Concept Training for Human-Aligned Language Models](https://drive.google.com/file/d/1Yxh7W-q4wvc99DgoxAilTM4bW12JZCwh/view?usp=drive_link)  | Christine Zhang  |  
| [Confusion-Set Guided Retrieval for LLM-Constrained Brain-to-Text Decoding](https://drive.google.com/file/d/1PV0Y2lYM_-nuynhFLBmhlS1x0CvOfZ7M/view?usp=drive_link)  | Andrew Su, Hyungjae Kim, Vincent Jinpeng Yip  |  
| [Context Under Pressure: How Language Model Agents Should Save and Read Information Over Long Interactions](https://drive.google.com/file/d/1itWG5cJUODG0_DHQmih85IeXUGJrqfeY/view?usp=sharing)  | Alexander Owen Worley  |  
| [Continuous Utility Direct Preference Optimization](https://drive.google.com/file/d/15LKUEvV6XkFR1Oo-dgcn06qIOy9hQghy/view?usp=drive_link)  | Muhammad Ahmed Mohsin  |  
| [Cost-Aware Escalation from Scalar Reward Models to Generative Models](https://drive.google.com/file/d/1wl8RMIzjdqTLO-T0w74IUR5xJ54Oqa8J/view?usp=sharing)  | Cole Yarbrough, Landon Renjiro Maka'ike Choy, Rui Chen  |  
| [Curriculum-Based Fine-Tuning for Summarization of Endometriosis Data](https://drive.google.com/file/d/1YysdJp7XdtAOLfmT_mZz9n7wfJKFkkgB/view?usp=sharing)  | Ali Hicham Tout  |  
| [Data-Centric Control of Verbosity for DPO-Based Instruction Alignment](https://drive.google.com/file/d/1fitHvha27RAwa_5CFMekHRbb011GF9RW/view?usp=sharing)  | Susan Lee, Will Richard Alex Furlow  |  
| [DeepRoot: Graph-Coordinated Multi-Agent Reasoning](https://drive.google.com/file/d/1Has1nwj6_-s4OY-6QaQiSxkG2p4RRnZG/view?usp=drive_link)  | Sean J Wang, Sijbren Manuel Kramer, Zijian (Carl) Ma  |  
| [Designing a Conservative Humor Filter Can a Model Tell If an Image Caption Is Funny?](https://drive.google.com/file/d/1BK3eoQSxe8dWu4B140uiqfRXI9tu1dXi/view?usp=drive_link)  | Michael Roger  |  
| [Diagnosing the Reversal Curse via Mechanistic Probing and Symmetric Training](https://drive.google.com/file/d/1Ohyv9v013IOuLjKY2ARra-FkvIfdH5SZ/view?usp=sharing)  | Deepti Gupta, Ke Huang, Rafael Cardoso Ferreira  |  
| [Diversity-Incentivized GRPO for Constrained Arithmetic Reasoning](https://drive.google.com/file/d/1VrZcxhIqy7lu6ZSmSrVokGx2YZJYvb_W/view?usp=drive_link)  | Gaurav Tyagi  |  
| [Do Language Models implement compositional solutions for natural language understanding](https://drive.google.com/file/d/1L3GK5XXXqDBhyMZLl3R0PXkzi0QzvftU/view?usp=drive_link)  | Ahmad Jabbar  |  
| [Do Long Contexts Help Legal Knowledge? A Case Study on US–China Securities Regulation](https://drive.google.com/file/d/1WWfuj6bKOuUykIzwMJt0AbCLtLAxzk81/view?usp=sharing)  | Yufei Peng  |  
| [Does Fine-Tuning Hurt Cross-Platform Generalization in Depression Detection?](https://drive.google.com/file/d/1cvKSQEc2-gHDU2AOA_bFBBZoQlKmfHeg/view?usp=sharing)  | Yanav Lall  |  
| [Does Pedagogy Hurt Truth? Evaluating Educational Rewriting in Medicine](https://drive.google.com/file/d/1MvqtDv2_qfkeDr3VI3_ouYxk4dSgM665/view?usp=sharing)  | Cally Lin, Sasa Simic  |  
| [Don’t Think About It: Activation Steering as Silent Defense Against Prompt Injection](https://drive.google.com/file/d/1tzfOGIeRZRTrcFS8HA-lwyWpDTnEIQrt/view?usp=drive_link)  | Gaurav Anand  |  
| [Dynamic Ledger: Retrieval-Augmented Structured Memory for Test-Time Learning](https://drive.google.com/file/d/1I6vW9WQnt6tM2iVQPQpEAxS68Z0-IhGK/view?usp=sharing)  | Jerry Gu, Sabrina Yen-Ko, Shurui Liu  |  
| [Dynamic Token Merging for Efficient Subword Encoder-Decoder Transformers](https://drive.google.com/file/d/1BXkobcZ22FagAsVXKMVEKggmwmcPSKaU/view?usp=drive_link)  | Chris Gu, Marco Andono Sie, Nathan Zhou  |  
| [Dynamic Token Merging for Encoder-Only Transformers: Adapting MrT5’s Delete Gate to BERT and XLM-RoBERTa](https://drive.google.com/file/d/1wKisWVnppkVUO-BRgmQUBZID55Oi4QtB/view?usp=drive_link)  | Aronima Dass, Hiva Zaad, Tianhui Huang  |  
| [Effect of Text Embedding Scale on GraphRAG Accuracy](https://drive.google.com/file/d/1ywk42ek3WHcQ8yO2oWkGyuVYWzWuUJnc/view?usp=sharing)  | Jon Valur Bjornsson  |  
| [Emotional Arc Preservation in LLM Literary Translation](https://drive.google.com/file/d/1IIzHl4n6Uzf801lUtVIXq1qJZzZhDdOs/view?usp=drive_link)  | Chloe Di Murdoch, Esidore Fajardo Eneinyang, Julia E Rhee  |  
| [End-to-End Driving Trajectory Prediction with Vision-Language-Action Model](https://drive.google.com/file/d/197_F2xFZNTCfY3dzfcy7PmF45-HkHAPa/view?usp=drive_link)  | Anze Liu  |  
| [Energy-Accuracy Trade-offs in Transformer-Based NLP Models: A Unified Benchmarking Study](https://drive.google.com/file/d/1wfnyuPQEvuHlv0425VAkuuHg9kfiIDcG/view?usp=sharing)  | Thibaud Xavier Clement  |  
| [Entropy-Triggered RAG: Optimizing Retrieval Efficiency via Token-Level Shannon Entropy](https://drive.google.com/file/d/1HAZquN1T5KpqnJBD-bAweH3M2_igxo9v/view?usp=drive_link)  | Yucheng Yao  |  
| [Evaluating Eliminative Reasoning in LLM-Based Differential Diagnosis](https://drive.google.com/file/d/1hEteXqvN7IwLJEJKn3Z1eugNSVlow9ZI/view?usp=drive_link)  | Seyun Bang, Tatiana Zhang  |  
| [Evaluating JEPA for Natural Language Tasks](https://drive.google.com/file/d/1zECH9eeSeygB2RUbeqH5cQV22BAZVv4x/view?usp=drive_link)  | Henry Jingsong Zhou, Oleh Ivankiv, Yousef Hassan Ramadan  |  
| [Evaluating Robustness of Large Language Models to Algospeak](https://drive.google.com/file/d/1TEjJATYSUyit32ZfaS2gUwYbjYZd_8uE/view?usp=drive_link)  | Hnin Yupar Mon, Thet Htar Thin Zar  |  
| [Evaluating Robustness of Social Bias Detection to Lexical and LLM-Driven Perturbations](https://drive.google.com/file/d/1LIkiEpnRqySU6XQ5MPP5SZ1o4t20UMNO/view?usp=drive_link)  | Nomin-Erdene Bayarsaikhan  |  
| [Evaluating User-Style Adaptation for Professional Text Generation](https://drive.google.com/file/d/11me1Cbb0vQrzuKf1RTwowJ2321hh2TxK/view?usp=drive_link)  | Andrea Ji Woo Nam Song  |  
| [FACT: Attention Consistency Training Mitigates Sycophancy and Jailbreaks](https://drive.google.com/file/d/1TxSWGpRRxNcInAHD6G4VLn0O8KDM7DjY/view?usp=drive_link)  | Emma Sampietro, Justin Nicolas Hartenstein  |  
| [FALCON: Factual-Aware Logical Consistency for Large Language Model Outputs via NLI-Guided Mixed Integer Optimization](https://drive.google.com/file/d/1D7vR-1porZQ_dXkCp4lJqQjFBtLG50-v/view?usp=sharing)  | Rehan Raza Azam  |  
| [Fast Compression versus Exact Recall: Investigating the Trade-offs Between Models in Specialized Reasoning Tasks](https://drive.google.com/file/d/1k3kz9b8qeOmivIhSm9AreNkybS5SfGbu/view?usp=drive_link)  | Jerry Xiao, Nick Yan  |  
| [Fast Vocabulary Transfer for African Languages in Multilingual Machine Translation](https://drive.google.com/file/d/1A79GQjv0BVaCqczSN0eptW1Pw7XD8AT_/view?usp=drive_link)  | Kailash Chandran Elumalai, Biya Brook  |  
| [From Premise to Punchline: A Fine-tuned Model and “Writers’ Room” Framework for Saturday Night Live Sketch Script Generation](https://drive.google.com/file/d/13CQkkS7UB2NF7jOJQFY2T0_rH_c7bJxy/view?usp=drive_link)  | Hannah Yu, Natalie Hampton  |  
| [From Private Memory to Collective Intelligence: Collaborative Test-Time Learning](https://drive.google.com/file/d/1K8x5HzguZkI7YRxovY9_rzd2utWa7sVz/view?usp=sharing)  | Jiaming Shen, Jiaxin Fang, Xinrui Jiang  |  
| [From Symptoms to Syndromes: Development and Validation of Fine-Tuning Transformer Architectures for Genetic Neurological Disease Diagnoses](https://drive.google.com/file/d/1_rqe5K68BlZfU9ngzXvXhtMDN-KxWY3K/view?usp=drive_link)  | Anushka Rawat, Ximing Gao, Yi Li  |  
| [Generative Dialogue State Tracking with GPT-2 for Task-Oriented Service Conversations](https://drive.google.com/file/d/12WXyOWHloAN_fabQFpWWyoJJtQITbmSH/view?usp=sharing)  | Venu Madhav Samprathi Ram Prasad  |  
| [Gold-Guided Programmatic Distillation for Financial](https://drive.google.com/file/d/1Ip2roz6ld4z5wVkzHxEJ55e543lQBFow/view?usp=sharing)  | Elana N Chen, Erica Zhao, Yun Dong  |  
| [Grounded Go Commentary Generation via Expert Engines and Structured Terminology](https://drive.google.com/file/d/14ZUN1BQUbRkSREBiM-gu8HHpq6xKJmyE/view?usp=drive_link)  | Yudong Chen  |  
| [HALLU-NLI: Revisiting Natural Language Inference (NLI) Hallucination Detection Methods forLLM-Generated Biographies](https://drive.google.com/file/d/1yNGtGUfLSVaVmoKd5ZRxkd9WWSfaRLpK/view?usp=drive_link)  | Nathania Elizabeth Lim, Sally Lee, Sarah Dong  |  
| [Hash Routed Delta Patches for Fast Knowledge Updates in Small LLMs](https://drive.google.com/file/d/1OTN52YfosGk6gWQV3rnukHiz8v9kpEqe/view)  | Arash Hamzehlou  |  
| [Hidden Signals: Hallucination Prediction in Medical QA](https://drive.google.com/file/d/10US7Q4aJ-RP7GwkpUknT9QpBrzPiwqnZ/view?usp=drive_link)  | Catherine M Zhang, Christina Ba, Ina Kathleen Chun  |  
| [How does fine-tuning change internal representations in an audio transcription model?](https://drive.google.com/file/d/1_4hMJiQpGl6fEgnk1uRXYbHZ9WjpmLsF/view)  | Brandon Liu, Jason Hu, Jenny Jin  |  
| [Improving Lean4 Autoformalization via Cycle Consistency Fine-tuning](https://drive.google.com/file/d/1E8AwwI3sasJ23rdtZTJYgoZaX16ptiz1/view?usp=drive_link)  | Arsen Shebzukhov  |  
| [Improving Scientific Reasoning in Small Language Models via Process Preference Re-Ranking](https://drive.google.com/file/d/1nB9sX-c-uHgmzZIGxKfPQuMXCmLnrOkj/view?usp=drive_link)  | Arya Gupta, Marianne Feng Liu  |  
| [Investigating the Impact of Persona-Based System Prompts During SFT of Code LLMs](https://drive.google.com/file/d/1SRnK5D_OPN_ywrlzRJ4eIHSzBj3uBjGo/view?usp=drive_link)  | Tushar Aggarwal  |  
| [Language-Augmented Flow Matching Policies for Robust Out-of-Distribution Robot Manipulation](https://drive.google.com/file/d/1QRgwykNRTOe9gjbedFdFPFRh85RrE3e6/view?usp=drive_link)  | Jeff Liu, Lucas Sosnick  |  
| [Language-Conditioned Objectives for Task-Agnostic Preference Learning and Controller Updates](https://drive.google.com/file/d/1ITpeEgnLdD_agdAVjiKdi5MoBH5p5PqF/view?usp=sharing)  | Kyeong-Won Park  |  
| [Learning a Discriminator for Conceptual Diversity in LM Outputs](https://drive.google.com/file/d/1Z54tr8YDfSD87DLkY30PtQSQO88UgZs_/view?usp=sharing)  | Hangoo Kang, James Liu  |  
| [Learning Efficient Tool Orchestration with Language Models](https://drive.google.com/file/d/14tRxzvU78ew8zU1N-vO3zZFaAQhqg-D_/view?usp=sharing)  | Orhun Akengin  |  
| [Learning from Critiques: A Geometric Framework for Response Improvement](https://drive.google.com/file/d/1uDmVqCqINye_PbcNTYM47O-5ypWzzBX_/view?usp=drive_link)  | Haozhan Gao  |  
| [Learning When to Speak: Teaching LLMs Silence Through Specialized DPO and Distillation](https://drive.google.com/file/d/1SYmlPjflriiNLH7MGFkHoRxD6sVVl8Jm/view?usp=drive_link)  | Allison Sara John, Anthony D Argyropoulos, Yubo Ruan  |  
| [Location, Location, Generation: Fine-Tuning a VLM for Real Estate Descriptions](https://drive.google.com/file/d/1aP_VhVcaBtByPzdEfbx24H72cAi6wN26/view?usp=sharing)  | Carey Chang, Niko Terebuh Ustin  |  
| [Measuring the Measure: Mechanistic Prompt Sensitivity for LLM-Based Populism Coding](https://drive.google.com/file/d/1M6DLRnTQWagJfoQWsXZsxs7fg8YbvYzE/view?usp=sharing)  | Jiehan Liu  |  
| [Mechanistic Deconvolution of Memory and Context in Quantum Language Models](https://drive.google.com/file/d/1B3czC7tI2weYRSE_tcfcQiYoLfNX40Y6/view?usp=sharing)  | Nathan Roll  |  
| [MedDistill: Improving Clinical LLM Performance Through Natural Language Tabular Insights](https://drive.google.com/file/d/1bHtlW6jec94guXB_AKDovgp8_MYdgc4w/view?usp=drive_link)  | Joshua Logan Shunk, Patrick Ruibin Li  |  
| [MGA: Mixed Gated Attention for Efficient Long Context Attention](https://drive.google.com/file/d/1vGdEqBQMUDpCAArRbkdfMSE1rgtBGA7Y/view?usp=sharing)  | Jen Ha, Bharat Kumar  |  
| [Mixture-of-Steering Vectors (MoSV): Sparse Gating for Compositional Hallucination Mitigation](https://drive.google.com/file/d/16oAcFWItUcuTHCKvJys_I4xrlquJJBNE/view?usp=drive_link)  | Daniel Winston Lee, Olufeolu Oluwapelumi Kolawole, Vedant Malolan Srinivas  |  
| [MoSA: Mixture-of-Specialized-Agents for Cost-Efficient Long-Document Question Answering](https://drive.google.com/file/d/1XzseLce2rEIht6j7oqrsXo3hTLbEiVzv/view?usp=sharing)  | Haseeb Ismail, Mert Karabiyik, Shayaan Memon  |  
| [Multi-Lane Retrieval-Augmented Generation for Pharmaceutical Regulatory Dossier Writing](https://drive.google.com/file/d/1wADVi0UQm6czisN7EWjcfuasWZSlViM-/view?usp=sharing)  | Omar Ingi Halldorsson  |  
| [MuTaP: Multi-Task Mutation Predictor via LoRA-Adapted ESM-2](https://drive.google.com/file/d/1C4gySk7Cv4k81NiGkIBU6tFmbDXHUFuu/view?usp=drive_link)  | Aya Aburous, Jad Bitar  |  
| [NanoVQA](https://drive.google.com/file/d/1tovOkZXR4tIO9Fo4-rn6X5PmY3AvyBtZ/view?usp=drive_link)  | Ellen Xu  |  
| [Non-Toxic Trash-Talking Fantasy Football](https://drive.google.com/file/d/1PbVjf12lZhRG3LRsvXzz81cv_Ml_7Xlk/view?usp=drive_link)  | Andrew Dana Lawlor, Xander William Russell  |  
| [On-Policy Context Distillation](https://drive.google.com/file/d/1xUK1r-dTEq7UBo7pj-b6aBF0MRXvnn6z/view?usp=sharing)  | Darynne Lee, Shizhe He, Simon Pritchard  |  
| [Perceptual-Aware Spatial Scene Synthesis (PASSS)](https://drive.google.com/file/d/1AH1JRzu9QfeBBGnM6oY6KWPXa4cDKzjS/view?usp=sharing)  | Karan Singh Soin, Na Young Son  |  
| [Pinpointing Latent Planning in Language Models with Lightweight Mechanistic Methods](https://drive.google.com/file/d/15Izvo-W26--FwJK-dc2ZQma_sFeIyk7K/view?usp=drive_link)  | Harshvardhan Singh, Nick Rui, Nicole Ma  |  
| [PocketSheet: Enhancing Test-Time Learning using Efficient Memory Augmentation in Small Language Models](https://drive.google.com/file/d/1NWpdGwd6_TO-s3Kx9SXPj0pE8Mo1YjAU/view?usp=sharing)  | Prabhjot Singh Rai, Sakthivel Sivaraman  |  
| [Practical and Interpretable Unfair ToS Detection: Comparing Legal-Bert, Linear Lexical Models, and Editable trees](https://drive.google.com/file/d/1fvLodRs0BzUA94a_5C0hhT3ffNUJeGWa/view?usp=sharing)  | Basel AlKanjo  |  
| [Practical Design Decisions Can Matter More Than Training Algorithm Choice: A Study of LLM-Based Rust Bug Repair](https://drive.google.com/file/d/1dVqfVL_LLGoa4CkUIvdgx_aVyqOvkVpn/view?usp=drive_link)  | Ethan Charles Morgan  |  
| [Precision Under Pressure: Pushing the Boundaries of the Accuracy-Efficiency Frontier in Question Answering with Mixture-of-Depths](https://drive.google.com/file/d/1kPMUsTnj4EpyLDsJiGNNDVVVM3CNpFjM/view?usp=drive_link)  | Haoyue Yang, Jan Miroslaw Kopanski, Soha Sultan  |  
| [Preference-Based Alignment of Code Generation for MCP Server Development](https://drive.google.com/file/d/1N-V0wHksHSzHkYDpUnFybXuXCEL_7yh6/view?usp=drive_link)  | Kristjan Dagur Egilsson, Rami Ratl Mrad  |  
| [Progressive Screenplay Narrative Understanding via Contrastive Learning](https://drive.google.com/file/d/1I9khhgynzDPWrkMsUUE-JGZZhzfdFonC/view?usp=drive_link)  | Luca Thomas Wheeler  |  
| [Quantized Pre-training for Small Mixture-of-Experts](https://drive.google.com/file/d/1OtLRIfesl9fUhTe5lygU0CCZ3laj-RhV/view?usp=drive_link)  | Raghavendra Pranith Koppula  |  
| [RAG-Based LLM Supported by Clinically Structured Re-Ranking, RL-Tuned Retrieval, and Agentic Workflow for ED Triage Prediction](https://drive.google.com/file/d/1cWrCv_QY1ZAH3IlwyPMIvHyA49NhyMOc/view?usp=drive_link)  | Charlotte Louise Kramer, Isha Arora, Nino Alex Triandafilidis  |  
| [Rapping in Role: A Study of Persona Robustness in Large Language Models](https://drive.google.com/file/d/18JbLpCRd6yLzI1RRHb0o9N7z93wdv6ET/view?usp=drive_link)  | Eunice Hyeyun Jung, Megan Ja  |  
| [Recursive Self-Improvement for Continual Adaptation in Code](https://drive.google.com/file/d/1zkARphNHWceBrX-5nLR1aNA7UwGBG7Pt/view?usp=drive_link)  | Aaditya Vikram Nalawade, Chandra Suda, Ethan David Goodhart  |  
| [Reward Design for Medical Safety: Reducing Sycophancy via Truth-Weighted RLHF](https://drive.google.com/file/d/1aPdKoqkBLaJ8gD0P23vHJ8NsEmI3osEX/view?usp=drive_link)  | Jillian Chang, Juli Huang, Michael Kuang Min Li  |  
| [Scaling Test-Time Compute to Improve Formal Reasoning in Lean via Compiler Feedback](https://drive.google.com/file/d/1I7v6oknMBHzf9XWd6cVg60CJPuAohvj7/view?usp=drive_link)  | Adam Joseph Banks, Alexander Huang  |  
| [Self-Distillation for Discrete Flow Map Consistency](https://drive.google.com/file/d/1z9AtkuYo-901Q2eLyFBqR4BJFabKNLDm/view?usp=sharing)  | Suchir Agarwal  |  
| [Small Models Think Big: Toward Effective Memory Distillation for Small Co-Scientists](https://drive.google.com/file/d/13IcNIBpHu1yB7j-_1sudxrDjJkBU2bwE/view?usp=sharing)  | Jaanak Prashar, Renn Su, Summer Olivia Royal  |  
| [Structural Line Markers and Multi-Pass Reranking for GPT-2 Sonnet Generation](https://drive.google.com/file/d/1N2HZKNBWWmhiEPmWYr065rx8LrPIabAu/view)  | Aalaap S Hegde, Mudit Baid, Rakshit Kaushik  |  
| [SUMMEHRY: LLMs for Generating Temporal Patient Vignettes](https://drive.google.com/file/d/1hTh86Bs20D92rpQU3Of9QqX32zSq9G8D/view?usp=sharing)  | Arlina Shen, Asmita Sood, Eashan Monga  |  
| [Support-Aware Retrieval of Evidence Passages for Community Notes](https://drive.google.com/file/d/1cwQVi2ea_TJxRj-qF4f29Q3FleqqmzN9/view?usp=drive_link)  | Dorian Scott Gulley, Dyllan Han  |  
| [SYMBRION: Symbol Context and Dream Ego Relations Across Lifelong Dream Series as a Tool for Psychoanalysis](https://drive.google.com/file/d/1hdBkEktCljkEFAoe9Z8WTKKj-F92sRuh/view?usp=sharing)  | Bobby Rohrkemper, Chia-Wei Cheng  |  
| [Test Time Training for Sample-Efficient Practical Molecular Optimization](https://drive.google.com/file/d/1ZZ-REn5Mx_ZaA2dOQxK56-kZk6Z2IK0U/view?usp=drive_link)  | Aaron Chee-Hung Lee, Ishvi Mathai  |  
| [Test-Time Training on Binary Sub-Problems](https://drive.google.com/file/d/1ZgFZvrKsLR8y1UGE96dQGhU0cpaiQsFQ/view?usp=drive_link)  | Andrew Sung, Darrow Robert Hartman, Leo Li  |  
| [The Efficiency Threshold: Few-Shot Prompting vs. LoRA](https://drive.google.com/file/d/1ysTrwt2BMOp51TKHqnNnJLEazBXPpJ00/view?usp=drive_link)  | Abi Lopez, Daniel Joseph Grossman, Shreyas Chikkanayakanahalli Seshadri  |  
| [The Feasibility of Token-Level Compute Allocation across Depth in Pretrained Transformers](https://drive.google.com/file/d/1zuk71MqsSW4_IQq9pK9NDzPNTRj3tfv8/view?usp=drive_link)  | Anjali Sreenivas, Yuchen Li  |  
| [The Rosetta Probe: Cross-Lingual Syntactic Transfer in Monolingual English BERT](https://drive.google.com/file/d/1rynQlpGQaqgrBAKQT2VJf0Dog5HmsxOo/view?usp=drive_link)  | Ananya Niharika Navale  |  
| [Towards Robust Natural-Language Proof Verification](https://drive.google.com/file/d/1IUIN_NbXi1BrTqNiKG97Mdrq5nNZoq1R/view?usp=drive_link)  | Slim Barkallah  |  
| [TRACE: Tool-augmented Reasoning via Atomic Cheatsheet Editing](https://drive.google.com/file/d/1iG2nPk_-tznWJohCS2C6MpHBHWDbZeh5/view?usp=sharing)  | Arnold Tianyi Yang, Kyleen Liao, Roshen Sanjay Nair  |  
| [Understanding Mechanisms of Sycophancy in Multi-turn Interactions](https://drive.google.com/file/d/13tnElrYc4-ubFVbsF_sy6UwetPPWBhme/view)  | Camila Blank  |  
| [Understanding Value Embeddings in GPT-2 Training Speedruns](https://drive.google.com/file/d/1P4JESBj5hSTMtDjublyD6WHojH2E1tew/view?usp=drive_link)  | Arihan Varanasi, Markus Zhang  |  
| [Verified Anchor Selection and Adaptive Curriculum for Dynamic Cheatsheet Memory](https://drive.google.com/file/d/1V-eXG68yC8VXe0AmUp6wThnT8MhA6PUk/view?usp=sharing)  | Mengqian Chen  |  
| [Verified On-Policy Self-Distillation](https://drive.google.com/file/d/1laPoglYuM5iTyTuqQUuHb0QYYIY7Qxov/view?usp=drive_link)  | Jack Li, Sophia Yinfan Li  |  
| [Verifier-Guided Reasoning for Cryptic Crossword Clue Solving](https://drive.google.com/file/d/1AeuzLcvm6t4ZSlpFYPdv7_V_p_hQdmv_/view?usp=drive_link)  | Aarav Arora, Caleb Youngjae Whang Choe, Shamit R Surana  |  
| [Visceral Judgment: LLM Refusal through Affective State](https://drive.google.com/file/d/1Mf8bsB_nt6orD6drDKaL_2Aaxb4DZ3KC/view?usp=drive_link)  | Nicolas Kennedy  |  
| [Vision-Language Model Router for Robotics](https://drive.google.com/file/d/1sFVPkY0YLhxLyUjfn-D4S5aBipSv9YNZ/view?usp=sharing)  | Jadelynn Kim Dao, Milan Ganai, Satvik Sharma  |  
| [Where Reasoning Branches: How Preference Pair Construction Shapes DPO for Mathematical Reasoning](https://drive.google.com/file/d/1Zg4lepccZfUX3nltyoiYHD6mzYeno-4u/view?usp=sharing)  | Duy Nguyen  |  
## Default Projects  
| Project name  | Authors  |  
| --- | --- |  
| [A Study of SFT-DPO Interaction and LoRA vs Full Fine-Tuning in Small Language Models](https://drive.google.com/file/d/1AJKaOE1Sd_FYSm7PUmiQ_Hr3JBHtXpPS/view?usp=sharing)  | Christy Yang, Yuming Feng  |  
| [Accelerated DPO Fine-tuning GPT-2 with Constructed Data](https://drive.google.com/file/d/1Duz7yR3nVvj5lhV2JdRWMcVUX28HOUoa/view?usp=sharing)  | Jessie Ou, Weixin Yu  |  
| [Accelerating Attention for GPT-2 Using FLASHATTENTION, Longformer, and cosFORMER](https://drive.google.com/file/d/1Laef75yBJxiepchqc4aptUHOSx3xpaub/view?usp=drive_link)  | Diego Sierra, Thomas Sarda, Tom-Eliot Jullien  |  
| [AdamW The Last LLM-Bender: The Legend of LoRA](https://drive.google.com/file/d/1QdtF2HEnIDvQz5lqNWDtm6x4P7dio8l1/view)  | Ari Barbella-Blaha, Kieran Javier Barrett  |  
| [Adapting GPT-2 for Sentiment Analysis, Paraphrase Detection, and Sonnet Generation](https://drive.google.com/file/d/1PbsUK0rpszCCFkdxKmkBrRLpWMOKQsUz/view?usp=sharing)  | Fiona Han, Samih Shaheen Qureshi, William Charles Rose  |  
| [Adapting GPT-2 Through Fine-Tuning Across NLP Tasks](https://drive.google.com/file/d/1GNSyY2A7-DabZxirsF2orF8QYdxCjcNz/view?usp=sharing)  | Ritu Patil  |  
| [Adapting Pretrained GPT-2 via LoRA: How Much Fine-Tuning Do We Actually Need?](https://drive.google.com/file/d/1Wg67tL6GnKz2D1wCsjLgZL5FvMVvMuG_/view?usp=sharing)  | Zengmingyu He, Zerong Chen  |  
| [Adaptive Mixture-of-Heads: Routing Attention Heads in GPT-2 with Fixed and Dynamic Sparsity](https://drive.google.com/file/d/1OoD36K9r2i64ljMGltXQ4MF-66iHAf2G/view?usp=drive_link)  | David Stutz, Ryder Fried  |  
| [An Investigation of GPT-2 Applications and Training Improvements, and Exploring Multi-Token Entity Predictions](https://drive.google.com/file/d/1W1YI8v6v0h7NxVpP3vkbSl8b88xtjf1e/view?usp=sharing)  | Ben Wengreen, Bhavya Ashish Shah, Jeffrey Meng  |  
| [Applying Direct Preference Optimization to Improve GPT-2 Sonnet Generation](https://drive.google.com/file/d/1HZICvgHHJQV5GaqbE9aZAzftKsd_0x3A/view?usp=drive_link)  | Aadhav Prabu  |  
| [Beyond Full Fine-tuning: Finding the Limits of GPT-2 Efficient Adaptation](https://drive.google.com/file/d/1a386BUpOIORNNkvZoUPAVwwJxkmwQJof/view?usp=drive_link)  | Alexander Huayi Zhong, Kaitlyn Angel Kwan, Songyu Han  |  
| [Build GPT-2](https://drive.google.com/file/d/17dg_LTOOwKa_Rn7QIwJh_Yu6zCksh4X7/view?usp=drive_link)  | Lucia Losada, Nicole Cortes  |  
| [Build GPT-2](https://drive.google.com/file/d/15EdKCWYKCu2Z-dmlcXEydv_UWrcfcFF2/view?usp=drive_link)  | Yuchan Guo, Yushi Feng  |  
| [Build GPT-2](https://drive.google.com/file/d/1nT4MpBMBrzUhLctFNh0bM0mpgPA8wn0e/view?usp=drive_link)  | Pengyu Mo, Shirley Yu, Yixiao Zhang  |  
| [Building GPT-2](https://drive.google.com/file/d/1-lDCeiS0bEbmNrhTwUofioAl97zSutiw/view?usp=sharing)  | Suzannah Dalton Wistreich  |  
| [Building GPT-2 and Perfecting Performance with Low-Rank Adaptation](https://drive.google.com/file/d/191108iPKBEo_GNNjPsME84wnn4k_nW65/view?usp=sharing)  | Chenyu Song, Juntao Cheng, Mingyang Li  |  
| [Building GPT-2 for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/1MdprVngstAHEEnLXatKr1IzOd6KCR2nW/view?usp=sharing)  | Yifan Guo  |  
| [Building GPT-2 with Finetuning Optimizations](https://drive.google.com/file/d/148dX_htuQWWH7lNWqkie_byL_Mb1piVx/view)  | Ethan R Lee, Ethan Y Lu, Jingyu Zhang  |  
| [Building GPT-2: Revisiting a Key Milestone of NLP](https://drive.google.com/file/d/1CIKDtFF_6uKXTw6S0l5fZE1veWa1-EaH/view?usp=sharing)  | Andy Tianqi Wang, Darren Chan, Derek Yan  |  
| [Circuit-Aware Analysis of LoRA Fine-Tuning: What Changes, Where, and Why?](https://drive.google.com/file/d/1MtCHz5iCtttdNAMAe3pKBxo3qu-oB_NQ/view?usp=sharing)  | Nathan Maidi  |  
| [Cloze-Style Paraphrase Detection and Sonnet Generation with GPT-2: Exploring LoRA and Decoding Strategies](https://drive.google.com/file/d/1jxG9xilR5lqjduCma6Liu5RZN5e46qkg/view?usp=sharing)  | Nick Fursa  |  
| [Co-Adaptation in LoRA: Target Placement Effects and Inter-Module Interactions in GPT-2](https://drive.google.com/file/d/1Fo5sZhV8QsgGmsmhsHYcnzTixu8_35AX/view?usp=drive_link)  | Shekhar Sharma  |  
| [Comparing ReFT and LoRA on Classification and Generative Tasks with GPT-2](https://drive.google.com/file/d/18YA5lhh5AX2J-q2FD1bFUdJ5gNSch3Kd/view)  | Ryan Patrick Catullo  |  
| [Cost–Performance Tradeoffs for GPT-2 Fine-Tuning: A Case Study on Paraphrase and Sonnet Continuation](https://drive.google.com/file/d/1QSztD3ShWSELirjuYv56fUHaiThAc2FA/view?usp=drive_link)  | Ricardo Ruiz  |  
| [CS 224N Default Project](https://drive.google.com/file/d/1pGNpJqHtRqn7N2EnOf7_gX8GasAWJd3O/view?usp=drive_link)  | Kayla Li, Yaojing Huang He  |  
| [Cutting Out the Middleman: Direct Preference Optimization for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/13XiM5EYzQTi-BLmNEbgLdQj74OR5A475/view?usp=drive_link)  | Justin Yuankai Leong, William Li  |  
| [Data Efficient Fine-Tuning and Alignment of GPT-2](https://drive.google.com/file/d/1dO6-2pd2Eki6O-kTVR4RaQY4498Khq3Q/view?usp=drive_link)  | Aryaman Gupta, Joseph Lee, Zeyuan Feng  |  
| [Default Final Project: Efficient Adaptiation of GPT-2 via LoRA](https://drive.google.com/file/d/1gZeX18ayyOWGPLZ3_UR-DTiQKi7QlrHL/view?usp=sharing)  | Pedro Gaspar Pires  |  
| [Direct Preference Optimization for Constrained Generation and Classification in GPT-2](https://drive.google.com/file/d/13xme8zHW1xPN2hAiQnz4GdcxBIb1bUWf/view?usp=sharing)  | Jingxiong Zhao, Weining Li  |  
| [Direct Preference Optimization for Improving Sonnet Generation](https://drive.google.com/file/d/1f669g6MmqVhClb_ABGlynE3EvHG3xQeV/view?usp=sharing)  | Gio Ty  |  
| [Direct Preference Optimization: From Paraphrase Detection to Sonnet Generation](https://drive.google.com/file/d/1-EVp65nVz5B-PsdaFsXxmFPupoCDWdmg/view?usp=sharing)  | Florencio Paucar Sedano  |  
| [Does the Optimizer Matter? LoRA vs Full Fine-Tuning in NLP](https://drive.google.com/file/d/1_22MJciMpiIVrS4HQCg6OdINnoEKFKuv/view?usp=drive_link)  | Andy Dimnaku  |  
| [DoRA the Explorer](https://drive.google.com/file/d/1D9WXwI4ir6Ea5rb6-msJyzmGjc1T6h3X/view?usp=sharing)  | Cayden Gu, Imogen Lee  |  
| [DoRA: Parameter-Efficient Fine-Tuning for GPT-2 on Cloze Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/1NecEyjIiCFZ0EchK9iYBgcjlaOZixtDd/view?usp=sharing)  | Aniket Gupta, Anjani Pangal, Mallika Parulekar  |  
| [DPO for Structural Sonnet Generation and Paraphrase Detection with GPT-2](https://drive.google.com/file/d/1JsMX52oaE6nU417teRref1h19Atg5JJG/view?usp=drive_link)  | Daniel Marcelo Mottesi, Diego Bustamante, Jason McLeod Amsler  |  
| [Effects of Quantization on GPT-2 Small](https://drive.google.com/file/d/1Ynj_c5vtnUSifFk7DcznTC_llS_oSmjW/view?usp=sharing)  | Isabella Lynne Jordan  |  
| [Efficiency and Inference: A Comparative Study of PEFT and Full Fine-Tuning](https://drive.google.com/file/d/1h7b47JeMTPyWEX3W51vzCbgcNI4FHT2u/view?usp=drive_link)  | Sanyam Gupta  |  
| [Efficiency in GPT-2: Parameter Adaptation, Quantization, and Synthetic Data Augmentation](https://drive.google.com/file/d/1gIl7dn4fYkZyGoqcY8mXhXoY8RlPIXbM/view?usp=sharing)  | Abhinav Chinta, Ethan Hersch, Ryan D'Cunha  |  
| [Efficiency–Performance Trade-offs in LoRA-family: Fine-Tuning Methods for GPT-2](https://drive.google.com/file/d/15t49xOHRtd_bIMigJBDhK_4IRuJgldK3/view?usp=drive_link)  | Christine Li, Jason Yan, Justin Li  |  
| [Efficient Adaptation and Structure-Aware Post-Training of GPT-2 for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/1x91I4_m3s_kYP14gWxLAr152DIYAf9zC/view?usp=drive_link)  | Brandon Michael Kunitzer, Koa Lanakila Chang  |  
| [Efficient Alignment Is All You Need](https://drive.google.com/file/d/1EJhx8kT4kHUyQ2beXv2GMQu7rIDfHChq/view?usp=drive_link)  | Lingbo Duan, Shatong Zhu, Yufei Liu  |  
| [Efficient Fine-Tuning and Alignment of GPT-2 for Downstream NLP Tasks](https://drive.google.com/file/d/1L8tIGSYlH_l23WY7FBc-H8LrH9slGG8a/view?usp=sharing)  | Adam Alhousiki, Kamal Mohammed ElMallah, Tommy Leong  |  
| [Efficient Fine-tuning of GPT-2 for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/1JzmjLHOy_w18yKdZJwBNsglK_y78sZbW/view?usp=sharing)  | Jonathan You  |  
| [Efficient Fine-Tuning of GPT-2 via Low-Rank Adaptation (LoRA)](https://drive.google.com/file/d/1jj4WN4fH2kVHtr7LeZ2XsVxZKj-R2Wtk/view?usp=drive_link)  | Min Zhang, Shang Gao, Shang Gao  |  
| [Efficient Fine-Tuning of GPT-2: LoRA, Hyperparameter Search, and Scaling for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/1EITdHRLtnI6H9YaD2o1jeeBN6-ez9dY-/view?usp=sharing)  | Brian Sha  |  
| [Efficient Steering and Preference Alignment: Applying LoReFT and DPO to a Custom GPT-2 Architecture](https://drive.google.com/file/d/1skQfJCQP0Rtf53RCBE9R9EChu9m2TEjo/view)  | Haonan Zhu  |  
| [Encoding Task Structure via Attention Biases and Adaptive Computation](https://drive.google.com/file/d/1Jnkw2SlJlMFOgP7cSlMj9p754fklENUn/view?usp=drive_link)  | Dario Gaitzi Soatto  |  
| [Enforcing Rigid Syntax: Using LoRA to Adapt GPT-2](https://drive.google.com/file/d/1pwBE0R8Png_moiITNM88X65-eJJL6yrM/view?usp=sharing)  | Monami Dutta Gupta  |  
| [Enhanced Hybrid Search for LLM Hyperparamter Optimization](https://drive.google.com/file/d/13aUcJVLa5GIYXTaLzv7SLC_0GuA8vVGs/view?usp=drive_link)  | Aaron Michael Sequeira, Avery Graham Voss, CJ Indart  |  
| [Evaluating LoRA for Efficient GPT-2 Fine-Tuning](https://drive.google.com/file/d/12XiTv71i6U9cVayPgK21aWaRCCdKpMW7/view?usp=sharing)  | Raymond Ruimeng Llata, Vania Chow  |  
| [Evaluating Low-Rank Adaptation and Nested Low-Rank Architectures for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/1amBW55C7ayeWiTi9ADsmx5VMv7YUsCZl/view?usp=sharing)  | Ian Yue-Ran Chen  |  
| [Evaluating Low-Rank Representation Finetuning for GPT-2 Downstream Tasks](https://drive.google.com/file/d/1E41mgI7tBiqsb6FomMYBEJXCklXIs4Wu/view?usp=drive_link)  | Alvin Ayuyo  |  
| [Evaluating Performance, Efficiency, and Memory Trade-offs in GPT-2 Attention Mechanisms](https://drive.google.com/file/d/1nka-XQczxHAcLEElfn5NljYTXrWrDbtE/view?usp=sharing)  | Devon Thomas Johnston Smith, Lily Annabelle Bailey  |  
| [Exploring decoding and efficiency strategies for GPT-2](https://drive.google.com/file/d/1wmGyMSv4DVU6eL25KD-wJWQcOq15z1ns/view?usp=drive_link)  | Stephanie Stephanie Vezich Tamayo  |  
| [Exploring LoRA Variants With GPT-2](https://drive.google.com/file/d/1hgAmdQ2m5mktzuYNI8sYoNeStl5jfQ8a/view?usp=drive_link)  | George Danchen Song, Justin Choo  |  
| [Exploring Low-Rank Adaptation for Efficient GPT-2 Fine-Tuning](https://drive.google.com/file/d/1mi9liL_WpNUUCtpJWN9UWZdtf_kGhLVW/view?usp=drive_link)  | Andy Zhang, Yi Lu  |  
| [Exploring Parameter-Efficient Fine-Tuning for Paraphrase Detection with GPT-2](https://drive.google.com/file/d/10KKYmGLBe1jTqRN5YRWG4z7vo0jbzELC/view?usp=sharing)  | Krisha K Chokshi  |  
| [Extending GPT-2 for Informal and Slang Aware Language Understanding](https://drive.google.com/file/d/1jBi3u2VPgi4NS4pmXZb7D3E5c9aJRQ_8/view?usp=drive_link)  | Dhruv Darshan Naik, Ruby Hernandez  |  
| [Fairness-Aware Fine-Tuning of GPT-2 for Paraphrase Detection](https://drive.google.com/file/d/1FmXex1CNP9NMM6Sn84QNOVNCnwjborKU/view?usp=sharing)  | Deonna Owens  |  
| [Fine-tuning GPT-2 for Sentiment Analysis, Paraphrase](https://drive.google.com/file/d/1xmlpbQJQxZrdG2_WxUC3A75KMKZwpGuO/view?usp=sharing)  | Liliana Carolina Santos-Deonizio  |  
| [Fine-Tuning GPT-2 for Sentiment Analysis, Paraphrase Detection, and Sonnet Generation with Parameter-Efficient Adaptation](https://drive.google.com/file/d/1J4W7b9v58gVCJKYY3Q2Pdl0JrGpCA0T0/view?usp=drive_link)  | Carl Liu, Zikun Zhu  |  
| [Fine-Tuning GPT-2 for Sentiment Analysis, Paraphrase Detection, Sonnet Generation and Political Affiliation Detection](https://drive.google.com/file/d/1D65CCiFxdQKmPb1TeWEXvY1IhcEt_vc0/view?usp=drive_link)  | Anna Wu, Iris Zixiao Xu, Samantha Malowane Leventis  |  
| [Fine-Tuning GPT-2 for Sentiment, Paraphrase, and Sonnet Tasks](https://drive.google.com/file/d/1FBo9BvfUCuCC20QQ0upQGsTmK9Gv6YZf/view?usp=drive_link)  | Walter Lopez Chavez  |  
| [Fine-tuning GPT-2 with LoRA](https://drive.google.com/file/d/1oQoKogYL85Effj5oZW4zwj1DoMUhhVz9/view?usp=drive_link)  | Manish Agarwal, Pierce Cailean Sayer Mullin  |  
| [Fine-tuning GPT-2 with LoRA and DPO for Accurate Classification and Constrained Generation](https://drive.google.com/file/d/1fqxfGOBKotJOp2LX12rubCsKePgtNKhe/view?usp=drive_link)  | Shaoxiong Zhang  |  
| [Fine-Tuning GPT-2: A Playground for Discriminative and Generative Adaptation Tasks](https://drive.google.com/file/d/17abn3dYR_DgIoUzdzgvjFF1l7VxBM2ni/view?usp=drive_link)  | Aditi Somayajula, Sahithi Ankireddy  |  
| [Fine-Tuning, Alignment, and Efficient Adaptation of GPT-2 for NLP Downstream Tasks](https://drive.google.com/file/d/10z59n1zmZNY6XH4n_3pHTqJj90TMrStz/view?usp=sharing)  | Ahmed Mohamed Hassan Khidre Elsherbiny, Izhan Hamza, Patrick Wang  |  
| [FlashAttention-Enhanced GPT-2 for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/1Q6Eb9Ejp8538UgBk3icjnqcyMdKAZeG3/view?usp=sharing)  | Katie Liu, Norah Asemota, William Yang  |  
| [From Detection to Generation: Fine-tuning Large GPT2 Models for Paraphrasing and Poetry](https://drive.google.com/file/d/1qGXHo8S6LzatnfjzfcIsNG_fw909v_Wz/view?usp=drive_link)  | Anna Guo  |  
| [From-Scratch GPT-2 and Efficient Adaptation](https://drive.google.com/file/d/1fuXq9natU2arO0jfl3J5KzlsIYF02Qbi/view?usp=drive_link)  | Bryan Alexis Pineda, Michael James Nixon  |  
| [Full Fine-Tuning v. LoRA: Parameter-Efficient Adaption of GPT-2 for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/1yMp8pUQ0EwNQunhtM7APptw8ea-2qhMZ/view?usp=sharing)  | Megha Bindiganavale, Rydham Goyal  |  
| [GaLore: Gradient Low-Rank Projection](https://drive.google.com/file/d/1QnHe6eA22mXQ4L2Yr8ya5FveWaqSWpr6/view?usp=sharing)  | Chung-Suen Stephen Chan  |  
| [GDPO for GPT-2](https://drive.google.com/file/d/1LochTTr3WNhrl3iISuoPLaT8cCp5sDqV/view?usp=sharing)  | Ahmed Sherif Ahmed Elbakry Mohamed  |  
| [GPT-2 Default Project with Attention-only LoRA for Paraphrase Detection](https://drive.google.com/file/d/154pEZ3qUrB1Tx7E4llZM0jXsUL5rHi3G/view?usp=sharing)  | Yiqing Liu  |  
| [GPT-2 Implementation and Speedup](https://drive.google.com/file/d/181U-jkKwmv4FKh3ddDCtDAXuUJ7RicDD/view?usp=drive_link)  | Alexia Huang, Qi Wu  |  
| [GPT-2 with LoRA Optimization](https://drive.google.com/file/d/19B_SZPaVneX_rtfQQ3s6VRsJRb9TccRo/view?usp=drive_link)  | Illia Shkirko, Janhavi Purkar, Zhang Bai-han  |  
| [GPT-2 with Varying Attention Mechanisms](https://drive.google.com/file/d/1cF37MNdJEZ9LYDp1p4RDnRwECm6kJC3Q/view?usp=sharing)  | Aneesh Akella  |  
| [GPT2 Optimization with PEFT and DPO](https://drive.google.com/file/d/16oGQ6iDz0A_gbrbNyb_YvQrOG3KlVHqq/view?usp=drive_link)  | Kiran Sun  |  
| [GRPOET-Rank: Group Relative Policy Optimization with External Text-Ranking](https://drive.google.com/file/d/1RViAU1BfMcIwxtBApgudYACIQFdNDZiI/view?usp=drive_link)  | Eric Liang, Jamin Jia-Ming Xie, William Z Liu  |  
| [Hardware-Aware Self-Attention for GPT-2: A FlashAttention-based Study](https://drive.google.com/file/d/1pNly8gig-XHGCwmjVt_UDHr6xVJ-I5PR/view?usp=sharing)  | Siri Garudanagiri Virupaksha  |  
| [Implementing a GPT-2 Decoder for Text Generation, Classification, and Paraphrase Detection](https://drive.google.com/file/d/1fEpy1TIXI_GYx7qyvG-cGR1qcXRur6iP/view?usp=sharing)  | Alma Oralia Minerva Cooper, Antra Nakhasi, Louis Weisdorf  |  
| [Implementing and Extending GPT-2 for Multi-Task NLP Applications: A Parameter-Efficient Fine-Tuning Perspective](https://drive.google.com/file/d/1efT1y-p5q1nWzlvPjYtgjPvRqy_iNDL6/view?usp=drive_link)  | Isabel Li, Lianyu Yao, Yunjie Xu  |  
| [Implementing and Fine-Tuning GPT2 for Sentiment Analysis, Paraphrase Detection, and Sonnet Generation](https://drive.google.com/file/d/1UNeprRXdQ5QLQ52y3TnQIwGruALR7rKD/view?usp=drive_link)  | Zhenghui Chen  |  
| [Improving GPT-2 Fine-Tuning through Parameter-Efficient Adaptation and Preconditioned Optimization](https://drive.google.com/file/d/1PoYHBN4zHBNasMqWMofaYvXGnd_Bxz_1/view?usp=drive_link)  | Akhilesh Varadan Balasingam, Georgios Mikos  |  
| [Improving GPT-2 Fine-Tuning with Direct Preference Optimization for Sonnet Generation and Paraphrase Detection](https://drive.google.com/file/d/1IBly-tyVFOiLfTmSHEGCouQ0hyDIHkEv/view?usp=drive_link)  | Anna Gutowska, Nicolas Bejar Arambula, Petru Cristian Budianu  |  
| [Improving GPT-2 with Reinforcement Learning from AI Feedback: Automated Judges for Aligned Sonnet Generation](https://drive.google.com/file/d/10jc58_fiVMpcxk3UPpJDFCF7ElaTJhLM/view?usp=drive_link)  | Jason Meng, Shinnosuke Yagi  |  
| [Improving Performance and Efficiency of GPT-2 on Sonnet Generation and Paraphrase Detection Tasks](https://drive.google.com/file/d/1ZzQ6xvldqkiTTDi6ogrgmpUNW5IcaCNm/view?usp=drive_link)  | Divya Bhojraj  |  
| [Investigating Structure Aware Decoding and Cloze Style Classification for Robust GPT 2 Fine Tuning](https://drive.google.com/file/d/1ojOgVCWmnmwBo1O06SNnnHxwrhPJ7Y-v/view?usp=drive_link)  | Anya Von Diessl  |  
| [KV Caching and Speculative Decoding at GPT-2 Scale: Acceptance, Cost Ratio, and the Limits of Speedup](https://drive.google.com/file/d/1Q4kw19wsL-EUwThLZnE5GdOS-5DrnTkf/view?usp=drive_link)  | Amulya Parthasarathy  |  
| [Lather, Rise, Repeat: The Shampoo Optimizer](https://drive.google.com/file/d/1qkhLQqPT45hgcCp0ngC1zedU5w967FAO/view?usp=sharing)  | Ricky Javier Rios  |  
| [Learning to Rhyme with Token-Weighted DPO](https://drive.google.com/file/d/1kjyeSbhgoQ2JCSlQcz3K50AZqcLNDUY6/view?usp=sharing)  | Fisher Marks  |  
| [Leveraging GPT-2 for Multiple Downstream NLP Tasks: Classification and Generation](https://drive.google.com/file/d/1mXnFCY57jDMPaVD2FJGpUG2FeInbcN5C/view?usp=sharing)  | Davi Ferreira Veronese  |  
| [Longformer-Style Sparse Attention for GPT-2](https://drive.google.com/file/d/1n_PYhYaB2cXrlPtTZA4NBa83cBciBpbP/view?usp=drive_link)  | Hoang D Nguyen, Peter Martin Alisky  |  
| [LoRA vs LoReFT: Parameter-Efficient Fine-Tuning of GPT-2](https://drive.google.com/file/d/1sv0roHyvDRzGXC5PUDPETmKU-WKN1Jh9/view?usp=sharing)  | Allen Yuan, Andrew Wooyong Chung, Ryan Joonwon Suh  |  
| [Lora-Enhanced GPT-2 with DPO for Sonnet Generation](https://drive.google.com/file/d/1-Mwt5eOvzYafoA6e2VKuL4uP4jOvxYlx/view?usp=sharing)  | Filip William Henriksson, Krish Maniar, Nicholas Simon Allen  |  
| [Low-Rank Adaptation and Preference Optimization for Accessible Multi-Task GPT-2 Fine-Tuning](https://drive.google.com/file/d/1nDdlonP1ycrhUSzRq8pRUSQ87_0cD3OY/view?usp=drive_link)  | Mahathi Mangipudi, Taylor Elizabeth Hamilton-Hankins, Tyler Kinh Ho  |  
| [Low-Rank Adaptation for Efficient GPT-2 Fine-Tuning: Evaluation Across Classification and Generation Tasks](https://drive.google.com/file/d/1Hx9GncSS7_zAmOPocePV-U_NH8FHddPf/view?usp=drive_link)  | Vivek Tiwari  |  
| [Low-rank fine-tuning of the GPT-2 model](https://drive.google.com/file/d/1Xujb_UuOq8Ut2D4OYLCXyniaotVwPg9l/view?usp=sharing)  | Rongge Yan  |  
| [Memory-Efficient Transformer Attention via Tiled FlashAttention-style Implementation](https://drive.google.com/file/d/1rbQk4d4RgUMLqMKc6nHE12GjBTD0J-tJ/view?usp=drive_link)  | Sagar Kapare  |  
| [Metric-Aligned Sonnet Generation with LoRA and Self-Critical RL Fine-tuning](https://drive.google.com/file/d/1Plwp8u6UKtc_NMd22PfnMnpu5DFXQEqw/view?usp=sharing)  | Timothy Yu, Xiang Wan  |  
| [MiniGPT: Implementation, Fine-Tuning, and Extensions for Constrained Generation](https://drive.google.com/file/d/1oKgrAUW491GGAkyWf4Ep78c1QV2dtLwW/view?usp=drive_link)  | Shiwei Que  |  
| [Modernizing GPT-2: Integrating Low-Rank Adaptation, FlashAttention, and Multi-Token Prediction for Efficient Sonnet Generation](https://drive.google.com/file/d/1879KD2fOI7RWftEGu-Jf9HmbQ9wA6BCf/view?usp=sharing)  | Manan Sheth, Sanjay Dixit Bhuvanagiri  |  
| [Optimizing GPT-2 for Downstream Tasks: An Exploration of PEFT, Preference Optimization, and SMART](https://drive.google.com/file/d/1iIl7bPC2BZDljTUDVnA-2hrSVHhrGIf9/view?usp=sharing)  | Anastasiya Masalava, Eva Casto, Michael Rybalkin  |  
| [Parameter-Efficient Adaptation of GPT-2 Across Classification and Generation Tasks](https://drive.google.com/file/d/14wPleW9N8-814edBkqxXhuk7_YBDHbkw/view?usp=sharing)  | Isaias Martinez, Kristine Ma, Varsha Saravanan  |  
| [Parameter-Efficient Adaptation of GPT-2 across Discriminative and Generative Tasks](https://drive.google.com/file/d/1iXhngbnHUOX3lI5x6jWjWWSeeN_Twn_-/view?usp=drive_link)  | Junran Jia, Xianya Fu  |  
| [Parameter-Efficient Fine-Tuning for GPT-2: Comparing LoRA and ReFT on Paraphrase Detection](https://drive.google.com/file/d/1gmW-sJjhdAn4gOqKMspfV2seJ4UXDwnF/view?usp=sharing)  | Matthias Jiro Walther, Ngoc Nguyen  |  
| [Parameter-Efficient Fine-Tuning of GPT-2 for Classification and Text Generation](https://drive.google.com/file/d/1iSSNcjx-jkWvjOFDePV78F-FJyCpFtjI/view?usp=drive_link)  | Ryan He  |  
| [Parameter-Efficient Fine-Tuning of GPT-2 Using DoRA](https://drive.google.com/file/d/1ZazFnTtPGASfltJUFczLOclT_8mQkWQz/view?usp=drive_link)  | Linika Goel, Mindy Kay Harkness  |  
| [Parameter-Efficient Fine-Tuning of GPT-2 using Low-Rank Adaptation (LoRA)](https://drive.google.com/file/d/1iHV_8_rbBWnBa7ktpM_G7IEAxxgLbywY/view?usp=sharing)  | Chris Alexander Perez  |  
| [Parameter-Efficient Fine-Tuning of GPT-2 with LoRA](https://drive.google.com/file/d/1GUJFS-dXmiUNqU-VrnJc_Fjjvv2dz16u/view?usp=drive_link)  | Chloe Yuri Jeon, Erick Angelo Ramirez  |  
| [Parameter-Efficient Fine-Tuning of GPT-2 with LoRA: A Systematic Study of Rank, Scale, and Learning Rate](https://drive.google.com/file/d/1yBsdqojRTLI4tKqd_CFIu8eIla5JrKSi/view?usp=sharing)  | Cuiyuanxiu Chen  |  
| [Parameter-Efficient Fine-Tuning of GPT-2: Comparing LoRA, LoReFT, and Prefix Tuning Across Classification and Generation Tasks](https://drive.google.com/file/d/1hVbdS5XrE4RRAIBknIDBNceIBZq6t6Sq/view?usp=drive_link)  | Sally Wang, Zijian Luo  |  
| [Parameter-efficient fine-tunings for Downstream Adaptation of GPT-2](https://drive.google.com/file/d/1LADUQoKryOemuOwFnuVWwqa4YuVBlFfn/view)  | Peiyu Li, Zefang Zhou  |  
| [Parameter-efficient Finetuning and Preference Optimization of GPT-2 for Downstream Tasks](https://drive.google.com/file/d/1ie-EE7LVhFo8RuAninUGvveUH0LpSPKT/view?usp=drive_link)  | Pankaj Rajak  |  
| [Paraphrase Detection and Sonnet Generation using GPT-2](https://drive.google.com/file/d/1zMaIyWB4F6rOhk7B0RPGHjFwprt2xokK/view?usp=sharing)  | Dongyu Jia, Omar Walid Ayoub  |  
| [Paraphrase Detection and Sonnet Generation with LoRA and DPO](https://drive.google.com/file/d/1g3zomSEpz3ELI2k-RDyBxnGplU6czJ1I/view?usp=sharing)  | Diya Bhattacharjee, Jaagat Prashar, Kyle Tianshi  |  
| [Preference Optimization for Parameter-Efficient Multi-Task Learning of GPT-2](https://drive.google.com/file/d/1ktuI8C3KtdYzWTlfFGx9ZGXjc0GfJYQO/view?usp=drive_link)  | Jiecong Tan, Mark Yang  |  
| [Preference-Optimized GPT-2 for Cloze-Style Paraphrase Detection and Sonnet Generation via DPO with GPT-Scored Pairs](https://drive.google.com/file/d/1bLgShbsqrv3CdnOpXka93ukBnTZyKQ-i/view?usp=sharing)  | Andrew Samuel Park, Arun J Moorthy, Welton T Wang  |  
| [Preference-Tuning GPT-2 with LoRA and DPO for Classification and Poetry](https://drive.google.com/file/d/1jl2zF7p7gb7fLcuqvtNu5fxaDm5iZ7So/view?usp=drive_link)  | Hiromichi Murakami, Yuliia Murakami  |  
| [QLoRA Fine-Tuning for Reliable Structured Tool Calls with GPT-2](https://drive.google.com/file/d/1p046Y-8ruVTrrTlpm48COf88HUFRBFGq/view?usp=sharing)  | Jay Khemchandani  |  
| [Quadapter: Adapter for GPT-2 Quantization](https://drive.google.com/file/d/1K82RinUwC_Ek4BqBpylhm07BJ_fI9U2_/view?usp=sharing)  | Ethan Cohen, Wesley Bian  |  
| [Quantization of GPT-2 for Running on Edge Devices](https://drive.google.com/file/d/1vf4Nl5DFSNNmX07S1fLr3GjfddRnSyI9/view?usp=sharing)  | Gordy D Sun  |  
| [Rank, Bits, and Data: Efficient GPT-2 Adaptation for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/1231EgbjkzDzRKBMQyLup7G4JmWJrm5ZZ/view?usp=drive_link)  | Kerui Lu, Silin Du  |  
| [Re-implementing GPT-2 for Classification, Paraphrase Detection, and Sonnet Generation](https://drive.google.com/file/d/1XzseLce2rEIht6j7oqrsXo3hTLbEiVzv/view?usp=sharing)  | Saanvi Reddy Thummalapally, Shani Su  |  
| [Robust Fine-Tuning of GPT-2 with SMART Regularization, LoRA, and Enhanced Decoding for Downstream NLP Tasks](https://drive.google.com/file/d/1aGFE4tDq6QtgHaD6dAIiFd8K4yBC638W/view?usp=drive_link)  | Jake Klosowski, Kevin Stephen, Alex E Wurm  |  
| [Robust GPT-2 Fine-Tuning via LoRA and Smoothness-Inducing Regularization](https://drive.google.com/file/d/1EzMI8lHi3agi3wiXmqXoaWc7RFA3J8tj/view?usp=drive_link)  | Pooya Nabavi  |  
| [SADPOSS G: Structure-Aware Direct Preference Optimization for Shakespearean Sonnet Generation](https://drive.google.com/file/d/1vpX_d08ISpOzWQyD9xquHqN8W-mLr8WB/view?usp=drive_link)  | Mario Felix Sumali  |  
| [Second-Order Optimization for GPT-2 Fine-Tuning: Exploring K-FAC for NLP Downstream Tasks](https://drive.google.com/file/d/1pd5dVrCb9Rf1hQmEIsSpfudoGSIpOqq0/view?usp=drive_link)  | Isabella Kai He  |  
| [Sentiment Classification using GPT-2 Representations](https://drive.google.com/file/d/1vkELkaxpxZib0WajgF_iF4ozrCsZ0aH9/view?usp=drive_link)  | Sahaj Saini  |  
| [SMART Regularization and DPO for GPT-2 Sonnet Generation](https://drive.google.com/file/d/1P8IXY6XlZdVYy2EVhNAMbjRxdPvc0yuE/view?usp=sharing)  | Mac Broido  |  
| [SMART-GPT2: Adapting SMART-Style Regularization for Decoder-Only Fine-Tuning](https://drive.google.com/file/d/1dmUNT-cI6HVfWw3lnX9F1qI2HwO3FP5A/view?usp=drive_link)  | Austin Ho  |  
| [SMART-GPT2: Adversarial Regularization for GPT-2 Fine-Tuning](https://drive.google.com/file/d/1HTOW5xlTJPTbrHNIhTKR0zYqBaL7pYLs/view?usp=drive_link)  | Bahram Y Mohmand, Noah Sabbavarapu, Zihan Wang  |  
| [SOAP and Sonnets: Improving the Optimization Efficiency of GPT-2](https://drive.google.com/file/d/1ZG6dFrBysozSoflnzMkTbluwDTAWaK_X/view?usp=sharing)  | Kenna Zeng  |  
| [Source, Relay, and Suppressor Heads in a Poetry Generation Circuit](https://drive.google.com/file/d/1UmXI40mLp8eXZb0yMIDLrou0snBXOVkl/view?usp=drive_link)  | Kai Wen, Shaoyi Zhang  |  
| [Sparse ReFT](https://drive.google.com/file/d/1Awpafs3rwnqCLVxUuKb6IDTZ-dLNW21F/view?usp=sharing)  | Jacob Daniel Householder  |  
| [SPLoRA: Sonnet Generation and Paraphrase Detection with LoRA](https://drive.google.com/file/d/1jb5Dwnv_oKO_JTyvYewh9JHd4CNTrPFx/view?usp=drive_link)  | Joseph Rabara Bailey, Maya Vendhan  |  
| [Streaming Sonnets: Efficient Generation with KV Caching and Quantization](https://drive.google.com/file/d/1HCz5V74TbNIFSGvtrnlKdzZhgx06BB2v/view?usp=drive_link)  | Codey Codey Sun, Michael Yang  |  
| [Style Steering of GPT-2 Sonnet Generation with DPO](https://drive.google.com/file/d/1s3NK95MicHU7F8asoj4c4Ut39ZoNhfvE/view?usp=drive_link)  | Claudia Perez D'Arpino  |  
| [Task-Dependent Effects of Parameter-Efficient Fine-Tuning: A GPT-2-Based Study](https://drive.google.com/file/d/181PkHbFEnmcEq5CKU0ip545FLZEPYcsi/view?usp=sharing)  | Weiwei Wu  |  
| [Task-Driven Fine-Tuning and Efficient Attention for GPT-2](https://drive.google.com/file/d/1j1OzGzjcIdTwj_mNewwrG_Urk1dnCr9R/view?usp=sharing)  | Sixian Du, Susan Li, Yuzhou Bian  |  
| [Task-Specific Fine-Tuning Strategies for Improving GPT-2 Across Classification and Generative NLP Tasks](https://drive.google.com/file/d/1gimQ6Q6UoRCw8afbPyJCbgbcsxxfamoH/view?usp=sharing)  | Austin Chen, Cheney Sang, Harris Alan Lee  |  
| [Task-Specific GPT-2 Adaptation: Structured LoRA for Paraphrase Detection and DPO for Sonnet Generation](https://drive.google.com/file/d/1jEFJUO_Y1tC1kmKxSv2XB3b7o01AyQXf/view?usp=sharing)  | Puyang Du, Xijia Liu  |  
| [TaskRank](https://drive.google.com/file/d/1aXggrpEDXHVWSK-D1bq-lqIb4MWMJ2ho/view?usp=sharing)  | Fabio Ibanez, Peter Jason Benitez  |  
| [Teaching LLMs to Forget Bad Data with Controlled Unlearning](https://drive.google.com/file/d/1nDZnosXhnvuHb3cZQIi7xSfIZqM8oDXW/view?usp=sharing)  | Hanyu Yang  |  
| [The LoRA(x)](https://drive.google.com/file/d/14G_kyXJGvpdKTA_VMcAOTK-u_-sqp2AZ/view?usp=drive_link)  | Gerwin Delsocora Mateo, Ryan Da  |  
| [Title of your project](https://drive.google.com/file/d/17uMYnDAoyJVy20Z_WcHBglE4lSSa9jHA/view?usp=sharing)  | Arun Brian Morris Chhetri, Ian Luka Lasic-Ellis, Marcus Batt Kushner  |  
| [Uncertainty-Aware Self-Training for Paraphrase Detection + Learned Reranking for Sonnet Generation](https://drive.google.com/file/d/17sSqs-R9V2aV_RKdw1dmzeKIjDKXrRLy/view?usp=sharing)  | Alex M Michael, Luis Marc Botin-Sanz de Sautuola, Xander Coulter Hnasko  |  
| [Weight Decomposition Matters: DoRA vs. LoRA for Small GPT-2 Task Adaptation](https://drive.google.com/file/d/1SG2kQlNx8C2v--7bw8yR-XID7fMNDin9/view?usp=sharing)  | Svea Drekshagen  |  
| [Where Does LoRA Actually Help? Probing Layer-Wise Adaptation in GPT-2 for Paraphrase Detection and Sonnet Generation](https://drive.google.com/file/d/152l05joOATOSuBQEe9-o02U8GioRTlvt/view?usp=drive_link)  | Mona Anvarihosseinabad  |