19 papers
Minimal Complexes of Rational Tangles via the Fukaya Category
Jonathan A. Higgins
We compute minimal complexes for oriented rational tangles, denoted , in terms of the topological Fukaya category of the -punctured disk…
Don't Just Look, Intervene: Perturbation Based Region Labeling for VQA Images
Marko Jojic, Zhaonan Li, Ben Zhou
Vision Language Models (VLMs) should rely on visual evidence that directly determines the correct answer, but supervision for grounding visual reasoning is often expensive to obtai…
Hindcast: Replaying Prediction Markets to Evaluate LLM Forecasters
Xiao Ye, Jacob Dineen, Evan Zhu +3
Forecasters are evaluated by backtesting, which replays resolved questions and grades the probability the system would have assigned before the outcome was known. For LLMs, two cha…
Robust Asynchronous Planning via Auto-Formalization
Jiayi Zhang, Jianing Yin, Ben Zhou +1
LLMs can plan by generating schedules directly as a Planner, or by translating a task into a formal specification for a solver as a Formalizer. Most real-world tasks are asynchrono…
Mid-Training with Self-Generated Data Improves Reinforcement Learning in Language Models
Aswin RRV, Jacob Dineen, Divij Handa +4
The effectiveness of Reinforcement Learning (RL) in Large Language Models (LLMs) depends on the nature and diversity of the data used before and during RL. In particular, reasoning…
Skill Reuse as Compression in Agentic RL
Zhikun Xu, Yu Feng, Jacob Dineen +3
Large language model agents trained with reinforcement learning (RL) often learn brittle, task-specific shortcuts. We hypothesize that agents generalize better when their successfu…
Vocabulary Dropout for Curriculum Diversity in LLM Co-Evolution
Jacob Dineen, Aswin RRV, Zhikun Xu +1
Co-evolutionary self-play, where one language model generates problems and another solves them, promises curriculum learning without human supervision. The promise breaks down earl…
Reliable Use of Lemmas via Eligibility Reasoning and SectionAware Reinforcement Learning
Zhikun Xu, Xiaodong Yu, Ben Zhou +6
Recent large language models (LLMs) perform strongly on mathematical benchmarks yet often misapply lemmas, importing conclusions without validating assumptions. We formalize lemma$…
Unbiased Visual Reasoning with Controlled Visual Inputs
Zhaonan Li, Shijie Lu, Fei Wang +11
End-to-end Vision-language Models (VLMs) often answer visual questions by exploiting spurious correlations instead of causal visual evidence, and can become more shortcut-prone whe…
CORE: Concept-Oriented Reinforcement for Bridging the Definition-Application Gap in Mathematical Reasoning
Zijun Gao, Zhikun Xu, Xiao Ye +1
Large language models (LLMs) often solve challenging math exercises yet fail to apply the concept right when the problem requires genuine understanding. Popular Reinforcement Learn…
Cognitive bias in LLM reasoning compromises interpretation of clinical oncology notes
Matthew W. Kenaston, Umair Ayub, Mihir Parmar +14
Despite high performance on clinical benchmarks, large language models may reach correct conclusions through faulty reasoning, a failure mode with safety implications for oncology…
Evaluating Medical LLMs by Levels of Autonomy: A Survey Moving from Benchmarks to Applications
Xiao Ye, Jacob Dineen, Zhaonan Li +11
Medical Large language models achieve strong scores on standard benchmarks; however, the transfer of those results to safe and reliable performance in clinical workflows remains a…
RELATE-Sim: Leveraging Turning Point Theory and LLM Agents to Predict and Understand Long-Term Relationship Dynamics through Interactive Narrative Simulations
Matthew Yue, Zhikun Xu, Vivek Gupta +3
Most dating technologies optimize for getting together, not staying together. We present RELATE-Sim, a theory-grounded simulator that models how couples behave at consequential tur…
RECAP: Transparent Inference-Time Emotion Alignment for Medical Dialogue Systems
Adarsh Srinivasan, Jacob Dineen, Muhammad Umar Afzal +3
Large language models in healthcare often produce emotionally flat or opaque responses, failing to provide the transparent reasoning required for clinical trust. We present RECAP (…
ThinkTuning: Instilling Cognitive Reflections without Distillation
Aswin RRV, Jacob Dineen, Divij Handa +4
Recent advances in test-time scaling have led to the emergence of thinking LLMs that exhibit self-reflective behaviors and multi-step reasoning. While RL drives this self-improveme…
CC-LEARN: Cohort-based Consistency Learning
Xiao Ye, Shaswat Shrivastava, Zhaonan Li +6
Large language models excel at many tasks but still struggle with consistent, robust reasoning. We introduce Cohort-based Consistency Learning (CC-Learn), a reinforcement learning…
AALC: Large Language Model Efficient Reasoning via Adaptive Accuracy-Length Control
Ruosen Li, Ziming Luo, Quan Zhang +4
Large reasoning models (LRMs) achieve impressive reasoning capabilities by generating lengthy chain-of-thoughts, but this "overthinking" incurs high latency and cost without commen…
BOW: Training Language Models to Reason Over Plausible Next Words
Ming Shen, Zhikun Xu, Jacob Dineen +2
Next-word prediction (NWP) trains language models against a single observed continuation, even though many contexts admit multiple plausible next words. Recent RL-based next-word r…
ToW: Thoughts of Words Improve Reasoning in Large Language Models
Zhikun Xu, Ming Shen, Jacob Dineen +6
We introduce thoughts of words (ToW), a novel training-time data-augmentation method for next-word prediction. ToW views next-word prediction as a core reasoning task and injects f…