PaperHub

Junyang Lin

~Junyang_Lin1

24
论文总数
12.0
年均投稿
5.9
平均评分
接收情况16/24
会议分布
ICLR
13
NeurIPS
6
ICML
4
COLM
1

发表论文 (24 篇)

202520

6.0
4

CARE: Decoding-Time Safety Alignment via Rollback and Introspection Intervention

NeurIPS 2025Poster
6.0
5

DataMan: Data Manager for Pre-training Large Language Models

ICLR 2025Poster
7.3
4

Chain of Execution Supervision Promotes General Reasoning in Large Language Models

NeurIPS 2025Poster
4.7
3

Disentangling Reasoning Tokens and Boilerplate Tokens For Language Model Fine-tuning

ICLR 2025withdrawn
6.6
4

MARGE: Improving Math Reasoning with Guided Exploration

ICML 2025Poster
5.8
5

Rotated Runtime Smooth: Training-Free Activation Smoother for accurate INT4 inference

ICLR 2025Poster
6.2
5

A Spark of Vision-Language Intelligence: 2-Dimensional Autoregressive Transformer for Efficient Finegrained Image Generation

ICLR 2025Poster
7.3
4

Parallel Scaling Law for Language Models

NeurIPS 2025Poster
4.6
5

Analyzing and Mitigating Inconsistency in Discrete Audio Tokens for Neural Codec Language Models

ICLR 2025Rejected
4.4
5

Rethinking Data Selection at Scale: Random Selection is Almost All You Need

ICLR 2025Rejected
6.3
3

Efficient Long Context Fine-tuning with Chunk Flow

ICML 2025Poster
3.5
4

Aligning Large Language Models via Self-Steering Optimization

ICLR 2025withdrawn
6.1
4

Synthesizing Software Engineering Data in a Test-Driven Manner

ICML 2025Poster
6.1
4

CateKV: On Sequential Consistency for Long-Context LLM Inference Acceleration

ICML 2025Poster
6.4
4

Teaching Language Models to Reason with Tools

NeurIPS 2025Poster
3.8
4

Language Models can Self-Lengthen to Generate Long Texts

ICLR 2025withdrawn
6.3
3

Self-Evolving Critique Abilities in Large Language Models

COLM 2025Poster
8.7
4

Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free

NeurIPS 2025Oral
7.8
4

Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning

NeurIPS 2025Poster
7.0
4

OpenHands: An Open Platform for AI Software Developers as Generalist Agents

ICLR 2025Poster