Himabindu Lakkaraju

Temporal Sparse Autoencoders: Leveraging the Sequential Nature of Language for Interpretability

Can Trustworthiness Generalize? Leveraging Weak Supervision for Stronger Models

EvoLM: In Search of Lost Language Model Training Dynamics

How Post-Training Reshapes LLMs: A Mechanistic View on Knowledge, Truthfulness, Refusal, and Confidence

More RLHF, More Trust? On The Impact of Preference Alignment On Trustworthiness

Inference-Time Reward Hacking in Large Language Models

Measuring the Faithfulness of Thinking Drafts in Large Reasoning Models

Follow My Instruction and Spill the Beans: Scalable Data Extraction from Retrieval-Augmented Generation Systems

Quantifying Generalization Complexity for Large Language Models

Towards Unifying Interpretability and Control: Evaluation via Intervention

On the Hardness of Faithful Chain-of-Thought Reasoning in Large Language Models

Weak-to-Strong Trustworthiness: Eliciting Trustworthiness with Weak Supervision

Generalized Group Data Attribution