Diyi Yang

Relative Scaling Laws for LLMs

Real-Time Reasoning Agents in Evolving Environments

Computer Agent Arena: Toward Human-Centric Evaluation and Analysis of Computer-Use Agents

How Dark Patterns Manipulate Web Agents

Optimas: Optimizing Compound AI Systems with Globally Aligned Local Rewards

AutoLibra: Agent Metric Induction from Open-Ended Human Feedback

GEM: A Gym for Generalist LLMs

The Ideation-Execution Gap: Execution Outcomes of LLM-Generated versus Human Research Ideas

Searching for Privacy Risks in LLM Agents via Simulation

Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration

AutoMetrics: Approximate Human Judgments with Automatically Generated Evaluators

ReplicationBench: Can AI Agents Replicate Astrophysics Research Papers?

OpenCUA: Open Foundations for Computer-Use Agents

Blackbox Model Provenance via Palimpsestic Membership Inference

The Unlearning Mirage: A Dynamic Framework for Evaluating LLM Unlearning

Information Retrieval Induced Safety Degradation in AI Agents

Internal Causal Mechanisms Robustly Predict Language Model Out-of-Distribution Behaviors

When Models Know More Than They Can Explain: Quantifying Knowledge Transfer in Human-AI Collaboration

Aligning Language Models with Demonstrated Feedback

Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers

No Preference Left Behind: Group Distributional Preference Optimization

SWE-bench Multimodal: Do AI Systems Generalize to Visual Software Domains?

Distilling an End-to-End Voice Assistant Without Instruction Training Data

Dynamic Skill Adaptation for Large Language Models