3 citations · 3 across the 13 of their papers we have counts for
15 papers
Playing Dumb: How a Chess Transformer Simulates Users Across Skill Levels
Difan Jiao, George Eilender, Zhenwei Tang +1
Generative models increasingly simulate users at different skill levels in education and AI evaluation. Yet it remains unclear whether lower-skill behavior reflects less internal k…
Agent Safety From Within: Detecting Harmful Trajectories from LLM Internal States
Difan Jiao, Ashton Anderson
Language model agents can now perform sophisticated sequences of actions via tools and harnesses, which has increased the scope of the damage they can cause. Guard models, however,…
Emulate or Estimate? The Divergent Strengths of Base and Post-Trained Language Models for Opinion Simulation
Seth Grief-Albert, Jessica Bo, Difan Jiao +1
Large language models are increasingly used to simulate human opinions, but prior work reports conflicting results: some studies find promising alignment with human survey data, wh…
Synthetic Persona Pretraining: Alignment from Token Zero
Julian Minder, Viktor Moskvoretskii, Raghav Singhal +12
As language-model-based AI is increasingly deployed in autonomous settings, aligning its goals and values with those of humans becomes critical. Today, alignment, and the assistant…
Tandem Reinforcement Learning with Verifiable Rewards
Difan Jiao, Raghav Singhal, Robert West +1
Reinforcement learning with verifiable rewards (RLVR) has driven substantial gains in language-model reasoning. However, models fine-tuned with RLVR need not remain compatible with…
SafeGEO: Understanding Generative Engine Optimization Risks in Recommendation Agents
Qianfeng Wen, Yifan Simon Liu, Xin Liu +4
Generative Engine Optimization (GEO) lets content owners rewrite web content to increase their visibility in generative systems. In recommendation agents, this creates a risk that…
MINER: Mining Multimodal Internal Representation for Efficient Retrieval
Weien Li, Rui Song, Zeyu Li +8
Visual document retrieval has become essential for accessing information in visually rich documents. Existing approaches fall into two camps. Late-interaction retrievers achieve st…
LLM Safety From Within: Detecting Harmful Content with Internal Representations
Difan Jiao, Yilun Liu, Ye Yuan +4
Guard models are widely used to detect harmful content in user prompts and LLM responses. However, state-of-the-art guard models rely solely on terminal-layer representations and o…
Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learning
Rohan Surana, Gagan Mundada, Xunyi Jiang +19
Reinforcement learning (RL) has become a central post-training tool for improving the reasoning abilities of large language models (LLMs). In these systems, the rollout, the trajec…
ThinkTwice: Jointly Optimizing Large Language Models for Reasoning and Self-Refinement
Difan Jiao, Qianfeng Wen, Blair Yang +2
We introduce ThinkTwice, a simple two-phase framework that jointly optimizes LLMs to solve reasoning problems and refine the answers, based on Group Relative Policy Optimization (G…
Understanding the Dynamics of Demonstration Conflict in In-Context Learning
Difan Jiao, Di Wang, Lijie Hu
In-context learning enables large language models to perform novel tasks through few-shot demonstrations. However, demonstrations per se can naturally contain noise and conflicting…
SEAM: Semantically Equivalent Across Modalities Benchmark for Vision-Language Models
Zhenwei Tang, Difan Jiao, Blair Yang +1
Evaluating whether vision-language models (VLMs) reason consistently across representations is challenging because modality comparisons are typically confounded by task differences…
Learning to Imitate with Less: Efficient Individual Behavior Modeling in Chess
Zhenwei Tang, Difan Jiao, Eric Xue +4
As humans seek to collaborate with, learn from, and better understand artificial intelligence systems, developing AIs that can accurately emulate individual decision-making becomes…
Maia-2: A Unified Model for Human-AI Alignment in Chess
Zhenwei Tang, Difan Jiao, Reid McIlroy-Young +3
There are an increasing number of domains in which artificial intelligence (AI) systems both surpass human ability and accurately model human behavior. This introduces the possibil…
SPIN: Sparsifying and Integrating Internal Neurons in Large Language Models for Text Classification
Difan Jiao, Yilun Liu, Zhenwei Tang +3
Among the many tasks that Large Language Models (LLMs) have revolutionized is text classification. Current text classification paradigms, however, rely solely on the output of the…