3 citations · 3 across the 16 of their papers we have counts for
18 papers
GeoShrink: Accelerating Diffusion Transformers with Two Lines of Code
Haosen Li, Wenshuo Chen, Shaofeng Liang +3
Diffusion transformers incur substantial inference cost through repeated model evaluations along a sampling trajectory. We introduce GeoShrink, a training-free acceleration method…
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression
DeepSeek-AI, :, Anyi Xu +585
The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work has substantially reduced the cost of long-context computation…
FlowAWR: Online Adaptive Flow Reinforcement via Advantage-Weighted Rectification
Zheming Fu, Ruizhe He, Wei Shang +5
Aligning generative flow models on continuous spaces via online reinforcement learning is constrained by intractable trajectory likelihoods. Existing density-approximated policy gr…
DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
DeepSeek-AI, Anyi Xu, Bangcai Lin +315
We present a preview version of DeepSeek-V4 series, including two strong Mixture-of-Experts (MoE) language models -- DeepSeek-V4-Pro with 1.6T parameters (49B activated) and DeepSe…
Diagnosing Harmful Continuation in Answer-Correct Long-CoT Training Traces
Chen He, Yuhao Wu, Lei Wang +2
Long chain-of-thought (CoT) traces are widely used as supervision for reasoning-oriented LLM SFT, yet answer-correct traces can still lead to markedly different fine-tuning outcome…
SKILLC: Learning Autonomous Skill Internalization in LLM Agents via Contrastive Credit Assignment
Hongxiang Lin, Zhirui Kuai, Erpeng Xue +1
Structured skill prompts improve exploration in long-horizon agentic reinforcement learning (RL). Skill-augmented RL methods retain external skills at inference, while skill-intern…
Detecting and Mitigating the Correct-Answer Extinction Window in Test-Time Reinforcement Learning with Majority Voting
Hongxiang Lin, Zhirui Kuai, Erpeng Xue +1
Test-time reinforcement learning (TTRL) reports substantial accuracy gains on mathematical reasoning benchmarks using majority vote as a pseudo-label signal. We argue these gains a…
Learning to Think in Physics: Breaking Shortcut Learning in Scientific Diffusion via Representation Alignment
Haozhe Jia, Pengyu Yin, Wenshuo Chen +6
Physics-informed diffusion models typically enforce PDE constraints only on final outputs, leaving intermediate representations unconstrained and prone to shortcut learning under s…
Scientific Logicality Enriched Methodology for LLM Reasoning: A Practice in Physics
Zhaoxin Yu, Nan Xu, Kun Chen +3
With the continuous advancement of reasoning abilities in Large Language Models (LLMs), their application to scientific reasoning tasks has gained significant research attention. C…
Oracle Noise: Faster Semantic Spherical Alignment for Interpretable Latent Optimization
Haosen Li, Wenshuo Chen, Lei Wang +3
Text-to-image diffusion models have achieved remarkable generative capabilities, yet accurately aligning complex textual prompts with synthesized layouts remains an ongoing challen…
-Sampling: Zero-Cost Zigzag Trajectories for Semantic Alignment in Diffusion Models
Haosen Li, Wenshuo Chen, Shaofeng Liang +3
Diffusion models have achieved unprecedented success in text-aligned generation, largely driven by Classifier-Free Guidance (CFG). However, standard CFG operates strictly on instan…
QaRL: Rollout-Aligned Quantization-Aware RL for Fast and Stable Training under Training--Inference Mismatch
Hao Gu, Hao Wang, Jiacheng Liu +9
Large language model (LLM) reinforcement learning (RL) pipelines are often bottlenecked by rollout generation, making end-to-end training slow. Recent work mitigates this by runnin…
MARS: Enabling Autoregressive Models Multi-Token Generation
Ziqi Jin, Lei Wang, Ziwei Luo +1
Autoregressive (AR) language models generate text one token at a time, even when consecutive tokens are highly predictable given earlier context. We introduce MARS (Mask AutoRegreS…
Scalable Chain of Thoughts via Elastic Reasoning
Yuhui Xu, Hanze Dong, Lei Wang +3
Large reasoning models (LRMs) have achieved remarkable progress on complex tasks by generating extended chains of thought (CoT). However, their uncontrolled output lengths pose sig…
A Minimalist Approach to LLM Reasoning: from Rejection Sampling to Reinforce
Wei Xiong, Jiarui Yao, Yuhui Xu +8
Reinforcement learning (RL) has become a prevailing approach for fine-tuning large language models (LLMs) on complex reasoning tasks. Among recent methods, GRPO stands out for its…
Reward Models Identify Consistency, Not Causality
Yuhui Xu, Hanze Dong, Lei Wang +2
Reward models (RMs) play a crucial role in aligning large language models (LLMs) with human preferences and enhancing reasoning quality. Traditionally, RMs are trained to rank cand…
MathHay: An Automated Benchmark for Long-Context Mathematical Reasoning in LLMs
Lei Wang, Shan Dong, Yuhui Xu +6
Recent large language models (LLMs) have demonstrated versatile capabilities in long-context scenarios. Although some recent benchmarks have been developed to evaluate the long-con…
ThinK: Thinner Key Cache by Query-Driven Pruning
Yuhui Xu, Zhanming Jie, Hanze Dong +6
Large Language Models (LLMs) have revolutionized the field of natural language processing, achieving unprecedented performance across a variety of applications. However, their incr…