LLM+XAI Reading Group

Upcoming Presentations

Past Presentations

• 2026-05-27 Tensor Product Representation Probes Reveal Shared Structure Across Linear Directions presented by Archit
• 2026-05-20 Interpreting CLIP with Sparse Linear Concept Embeddings (SpLiCE) presented by Jae
• 2026-05-13 How Emotion Shapes the Behavior of LLMs and Agents: A Mechanistic Study presented by Burak
• 2026-05-06 Can VLMs Actually See and Read? A Survey on Modality Collapse in Vision-Language Models presented by Sergio
• 2026-04-01 Less is More: Recursive Reasoning with Tiny Networks presented by Archit
• 2026-03-25 Monitoring Emergent Reward Hacking During Generation via Internal Activations presented by Wenhao
• 2026-03-18 From Directions to Regions: Decomposing Activations in Language Models via Local Geometry presented by Thomas
• 2026-03-11 LatentLens: Revealing Highly Interpretable Visual Tokens in LLMs presented by Jae
• 2026-02-25 Think Deep, Not Just Long: Measuring LLM Reasoning Effort via Deep-Thinking Tokens presented by Burak
• 2026-02-18 Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs presented by Hassan
• 2026-02-11 Steering Llama 2 via Contrastive Activation Addition presented by Sergio
• 2026-01-21 X-Fusion: Introducing New Modality to Frozen Large Language Models presented by Archit
• 2025-12-10 ReFT: Representation Finetuning for Language Models presented by Thomas
• 2025-12-03 Multi-Turn Puzzles: Evaluating Interactive Reasoning and Strategic Dialogue in LLMs presented by Wenhao
• 2025-11-26 RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning and Verification in Long-Horizon Generation presented by Hassan
• 2025-11-12 FaCT: Faithful Concept Traces for Explaining Neural Network Decisions presented by Sergio
• 2025-11-05 Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas presented by Jae
• 2025-10-29 Enhancing Chain of Thought Prompting in Large Language Models via Reasoning Patterns presented by Burak
• 2025-06-25 π0.5: a Vision-Language-Action Model with Open-World Generalization presented by Kun
• 2025-06-18 Reasoning Models Don't Always Say What They Think presented by Hergen
• 2025-06-11 LLMs are Greedy Agents: Effects of RL Fine-tuning on Decision-Making Abilities presented by Wenhao
• 2025-06-04 LLMs Get Lost in Multi-Turn Conversation presented by Burak
• 2025-05-28 Do Language Models Use Their Depth Efficiently? presented by Jae
• 2025-05-21 Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketching presented by Hassan
• 2025-05-14 Your Large Language Models Are Leaving Fingerprints presented by Sergio
• 2025-04-02 Robotouille: An Asynchronous Planning Benchmark for LLM Agents presented by Hassan
• 2025-03-26 Physics of Language Models: Part 3.1 presented by Hergen
• 2025-03-12 Learning to Reason at the Frontier of Learnability presented by Wenhao
• 2025-02-19 The Geometry of Concepts: Sparse Autoencoder Feature Structure presented by Jeremy
• 2025-02-12 LaMI: Large Language Models for Multi-Modal Human-Robot Interaction presented by Sergio
• 2025-02-05 DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning presented by Xufeng
• 2025-01-29 Can LLMs Learn from Previous Mistakes? Investigating LLMs' Errors to Boost for Reasoning presented by Burak
• 2025-01-15 Language Models Represent Space and Time presented by Jae
• 2024-12-18 Deploying and Evaluating LLMs to Program Service Mobile Robots presented by Hassan
• 2024-12-11 Interpreting Emergent Planning in Modelfree Reinforcement Learning presented by Wenhao
• 2024-12-04 Embodied Agent Interface: Benchmarking LLMs for Embodied Decision Making presented by Xiaowen
• 2024-11-27 Towards More Faithful Natural Language Explanation Using Multi-Level presented by Sergio
• 2024-11-20 Sparse Crosscoders for Cross-Layer Features and Model Diffing presented by Jeremy
• 2024-10-30 GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models presented by Hassan
• 2024-10-23 Robotic Control via Embodied Chain-of-Thought Reasoning presented by Wenhao
• 2024-10-16 AffordanceLLM: Grounding Affordance from Vision Language Models presented by Xiaowen
• 2024-10-09 Interpreting Attention Layer Outputs with Sparse Autoencoders presented by Sergio
• 2024-09-25 A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and Toxicity presented by Xintong
• 2024-09-11 Understanding Social Reasoning in Language Models with Language Models presented by Burak
• 2024-09-04 Do Large Language Models Latently Perform Multi-Hop Reasoning? presented by Jae
• 2024-08-21 A Multimodal Automated Interpretability Agent presented by Wenhao
• 2024-08-14 A Concept-Based Explainability Framework for Large Multimodal Models presented by Jae
• 2024-08-07 Advancing LLM Reasoning Generalists with Preference Trees presented by Hassan
• 2024-07-31 STaR: Self-Taught Reasoner presented by Imran
• 2024-07-24 Neuron to Graph: Interpreting Language Model Neurons at Scale presented by Sergio
• 2024-07-17 Detecting hallucinations in large language models using semantic entropy presented by Xiaowen
• 2024-07-10 Identifying Linear Relational Concepts in Large Language Models presented by Jae
• 2024-06-26 What does the Knowledge Neuron Thesis Have to do with Knowledge presented by Xintong
• 2024-06-19 LLMs with Chain-of-Thought Are Non-Causal Reasoners presented by Wenhao
• 2024-06-12 Self-Supervised Alignment with Mutual Information: Learning to Follow Principles without Preference Labels presented by Xufeng
• 2024-06-05 A Survey on Evaluation of Large Language Models presented by Hassan
• 2024-05-29 Inference-Time Intervention: Eliciting Truthful Answers from a Language Model presented by Sergio
• 2024-05-22 Finding Neurons in a Haystack: Case Studies with Sparse Probing presented by Imran
• 2024-05-15 Towards Uncovering How Large Language Model Works: An Explainability Perspective presented by Jae