10 papers
Branch2Skill: Efficient Skill Evolution Through Reasoning Trees
Yanwei Ren, Haotian Zhang, Likang Xiao +5
Skill evolution improves agent skills through feedback over time, with failed trajectories often providing informative signals by revealing incomplete or misleading behaviors. Howe…
Recycling Failures: Salvaging Exploration in RLVR via Fine-Grained Off-Policy Guidance
Yanwei Ren, Haotian Zhang, Likang Xiao +6
Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a powerful paradigm for enhancing the complex reasoning capabilities of Large Reasoning Models. However, standa…
Remodeling Semantic Relationships in Vision-Language Fine-Tuning
Xiangyang Wu, Liu Liu, Baosheng Yu +2
Vision-language fine-tuning has emerged as an efficient paradigm for constructing multimodal foundation models. While textual context often highlights semantic relationships within…
SPOGW: a Score-based Preference Optimization method via Group-Wise comparison for workflows
Yitong Cui, Liu Liu, Baosheng Yu +5
Large language models (LLMs) have exhibited significant capabilities in addressing challenging problems throughout various fields, often through the use of agentic workflows that a…
ContextPRM: Leveraging Contextual Coherence for multi-domain Test-Time Scaling
Haotian Zhang, Liu Liu, Baosheng Yu +5
Process reward models (PRMs) have demonstrated significant efficacy in enhancing the mathematical reasoning capabilities of large language models (LLMs) by leveraging test-time sca…
Re-Initialization Token Learning for Tool-Augmented Large Language Models
Chenghao Li, Liu Liu, Baosheng Yu +2
Large language models have demonstrated exceptional performance, yet struggle with complex tasks such as numerical reasoning, plan generation. Integrating external tools, such as c…
Instruction Learning Paradigms: A Dual Perspective on White-box and Black-box LLMs
Yanwei Ren, Liu Liu, Baosheng Yu +2
Optimizing instructions for large language models (LLMs) is critical for harnessing their full potential in complex and diverse tasks. However, relying solely on white-box approach…
LARGO: Low-Rank Regulated Gradient Projection for Robust Parameter Efficient Fine-Tuning
Haotian Zhang, Liu Liu, Baosheng Yu +3
The advent of parameter-efficient fine-tuning methods has significantly reduced the computational burden of adapting large-scale pretrained models to diverse downstream tasks. Howe…
SIGMA: Refining Large Language Model Reasoning via Sibling-Guided Monte Carlo Augmentation
Yanwei Ren, Haotian Zhang, Fuxiang Wu +4
Enhancing large language models by simply scaling up datasets has begun to yield diminishing returns, shifting the spotlight to data quality. Monte Carlo Tree Search (MCTS) has eme…
Generated Contents Enrichment
Mahdi Naseri, Jiayan Qiu, Zhou Wang
We study Generated Contents Enrichment (GCE), a conditional image-generation task in which a sparse scene description is first enriched through an explicit scene representation and…