Xiao Luo

Supervised fine-tuning (SFT) is fundamental to adapting large language models, yet training on complete datasets incurs prohibitive costs with diminishing returns. Existing data selection methods suffer from severe domain specificity: techniques optimized for general instruction-following fail on...

Reinforcement learning (RL) has emerged as a powerful post-training paradigm for enhancing the reasoning capabilities of large language models (LLMs). However, reinforcement learning for LLMs faces substantial data scarcity challenges, including the limited availability of high-quality external s...

Mathematical reasoning is one of the core capabilities for Large Language Models (LLMs). Yet, existing approaches often rely on static heuristics or pre-determined reasoning strategies, limiting their ability to adapt to different intermediate states. To address this limitation, we propose FLAIR...

Large language models (LLMs) are increasingly applied to scientific research, yet existing evaluations often fail to reflect the fine-grained capabilities required in practice. Most benchmarks are manually curated or domain-generic, limiting scalability and alignment with real scientific use case...

Self-training has emerged as a promising direction for autonomously improving large language models (LLMs). Existing approaches typically adopt a \textit{generate-and-filter} paradigm based on rejection sampling, which could suffer from inefficiency and low-quality reasoning paths. Towards this e...

Previous LLMs-based RL studies typically follow either supervised learning with high annotation costs, or unsupervised paradigms using voting or entropy-based rewards. However, their performance remains far from satisfactory due to the substantial annotation cost and issues such as model collapse...

Multi-step retrieval-augmented generation has attracted increasing attention due to its capacity to improve the factuality of large language models with iterative retrieved knowledge. However, the performance of multi-step RAG systems is susceptible to potential retrieval noise and fabricated doc...

This paper studies the problem of test-time adaptation for vision-language models (VLMs). Recent approaches typically measure the prediction entropy to store a confident cache for logit refinement. However, these confident samples tend to approach prototypes with limited coverage of data distribu...

Retrieval-augmented generation (RAG) is indispensable for enhancing large language models. However, RAGs are increasingly susceptible to poisoning attacks, in which adversarial documents are injected to manipulate generator outputs. Previous methods rely on output-side signals such as perplexity...

Retrieval-augmented generation (RAG) is widely adopted for radiology report generation with medical vision-language models, leveraging external reports as linguistic references. However, existing RAG methods rely primarily on dense embedding similarity, which may retrieve reports that are semanti...