1 citations · 2 across the 10 of their papers we have counts for
14 papers
Questioning the Questions: Sustaining Self-Evolution in Reasoning Models
Jinyuan Li, Chengsong Huang, Langlin Huang +4
Self-evolving reasoning models learn from their own generated questions, yet repeated self-training can lead to performance collapse. In this paper, we investigate why performance…
On the Off-Policy Teacher in On-Policy Distillation
Langlin Huang, Hao Liu, Mononito Goswami +6
On-policy distillation (OPD) has recently emerged as a promising post-training paradigm in which the student learns from trajectories generated by its own policy under dense teache…
MILO: Automated Harness Discovery via Orchestrated Multi-Agent Evolution
Prithwish Jana, Mononito Goswami, Hao Liu +9
Modern agentic systems combine an AI model with a harness that controls execution and environmental interactions. Harness design strongly affects long-horizon performance, yet its…
Hermes: Learning Contextual Reasoning Unlocks Test-Time Scaling
Xinyu Li, Mononito Goswami, Hao Liu +5
Test-time scaling improves model performance by allocating additional compute during inference. Using this compute effectively across multiple context windows requires deciding how…
Process Rewards with Learned Reliability
Jinyuan Li, Langlin Huang, Chengsong Huang +5
Process Reward Models (PRMs) provide step-level feedback for reasoning, but current PRMs usually output only a single reward score for each step. Downstream methods must therefore…
G-Zero: Self-Play for Open-Ended Generation from Zero Data
Chengsong Huang, Haolin Liu, Tong Zheng +7
Self-evolving LLMs excel in verifiable domains but struggle in open-ended tasks, where reliance on proxy LLM judges introduces capability bottlenecks and reward hacking. To overcom…
Nonsense Helps: Prompt Space Perturbation Broadens Reasoning Exploration
Langlin Huang, Chengsong Huang, Jinyuan Li +3
Reinforcement learning with verifiable rewards, particularly Group Relative Policy Optimization (GRPO), has significantly advanced the reasoning capabilities of Large Language Mode…
Training Data Efficiency in Multimodal Process Reward Models
Jinyuan Li, Chengsong Huang, Langlin Huang +4
Multimodal Process Reward Models (MPRMs) are central to step-level supervision for visual reasoning in MLLMs. Training MPRMs typically requires large-scale Monte Carlo (MC)-annotat…
LiViBench: An Omnimodal Benchmark for Interactive Livestream Video Understanding
Xiaodong Wang, Langling Huang, Zhirong Wu +4
The development of multimodal large language models (MLLMs) has advanced general video understanding. However, existing video evaluation benchmarks primarily focus on non-interacti…
POSS: Position Specialist Generates Better Draft for Speculative Decoding
Langlin Huang, Chengsong Huang, Jixuan Leng +2
Speculative decoding accelerates Large Language Model (LLM) inference by using a small draft model to predict multiple tokens, and a large target model to verify these tokens in pa…
CrossWordBench: Evaluating the Reasoning Capabilities of LLMs and LVLMs with Controllable Puzzle Generation
Jixuan Leng, Chengsong Huang, Langlin Huang +4
Existing reasoning evaluation frameworks for Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) predominantly assess either text-based reasoning or vision-langua…
Efficient Test-Time Scaling via Self-Calibration
Chengsong Huang, Langlin Huang, Jixuan Leng +2
Increasing test-time computation is a straightforward approach to enhancing the quality of responses in Large Language Models (LLMs). While Best-of-N sampling and Self-Consistency…
Divide, Reweight, and Conquer: A Logit Arithmetic Approach for In-Context Learning
Chengsong Huang, Langlin Huang, Jiaxin Huang
In-Context Learning (ICL) emerges as a key feature for Large Language Models (LLMs), allowing them to adapt to new tasks by leveraging task-specific examples without updating model…
Towards reporting bias in visual-language datasets: bimodal augmentation by decoupling object-attribute association
Qiyu Wu, Mengjie Zhao, Yutong He +4
Reporting bias arises when people assume that some knowledge is universally understood and hence, do not necessitate explicit elaboration. In this paper, we focus on the wide exist…