9 papers
Ladders of Thought: A Self-Evolving Curriculum of Progressively Simplified Reasoning Traces
Minghui Liu, Thomas Magelinski, Dehao Yuan +2
Large language models (LLMs) excel at reasoning when scaled to hundreds of billions of parameters, but small- and mid-scale models remain brittle reasoners even with knowledge dist…
When AI Reviews Train AI Reviewers: Scientific-Judgment Collapse and Mitigation
Sy-Tuyen Ho, Minghui Liu, Furong Huang +1
LLMs increasingly assist scientific peer review. This creates a feedback loop when generated reviews enter the training data of future LLMs. While recursive training has been studi…
-OPSD: Deriving with Policy Optimization, Training with Self-Distillation
Jiawei Xu, Minghui Liu, Juzheng Zhang +2
On-policy self-distillation (OPSD) is a promising approach to improve reasoning language models, but it remains brittle in practice: making it work reliably often requires substant…
Scheduling Thoughts: Learning the Order of Thought in Diffusion Language Models
Jiawei Xu, Minghui Liu, Aakriti Agrawal +2
Masked diffusion language models decode by iteratively unmasking tokens, where the unmasking order defines an "order of thought" that strongly influences generation quality yet is…
VeriGate: Verifier-Gated Step-Level Supervision for GRPO
Aakriti Agrawal, Minghui Liu, Furong Huang
Group Relative Policy Optimization (GRPO) is an effective recipe for training reasoning models with verifier-based outcome rewards, but its supervision is sparse: when all sampled…
SoundnessBench: Can Your AI Scientist Really Tell Good Research Ideas from Bad Ones?
Sy-Tuyen Ho, Minghui Liu, Huy Nghiem +6
Autonomous AI research agents aim to speed up scientific discovery by automating the research pipeline from hypothesis generation to peer review. However, existing benchmarks rarel…
Agentic Critical Training
Weize Liu, Minghui Liu, Sy-Tuyen Ho +5
Imitation learning (IL) teaches language-model agents to reproduce expert actions but not to distinguish them from plausible mistakes. Self-reflection methods expose models to alte…
Hold Onto That Thought: Assessing KV Cache Compression On Reasoning
Minghui Liu, Aadi Palnitkar, Tahseen Rabbani +9
Large language models (LLMs) have demonstrated remarkable performance on long-context tasks, but are often bottlenecked by memory constraints. Namely, the KV cache, which is used t…
MORSE-500: A Programmatically Controllable Video Benchmark to Stress-Test Multimodal Reasoning
Zikui Cai, Andrew Wang, Anirudh Satheesh +10
Despite rapid advances in vision-language models (VLMs), current benchmarks for multimodal reasoning fall short in three key dimensions. First, they overwhelmingly rely on static i…