影响力指数

74.73/100

前 1.9%

全站排名 #1,235

发表论文14 篇

平均评分5.6

年均产出4.7 篇/年

Gabriel Synnaeve

Research Scientist@Facebook·法国·OpenReview

研究方向

Large Language Models · Code generation · Reinforcement learning · Deep learning (convnets · RNNs...) · Vision (image recognition / object detection) · Speech/language technologies · Game AI

5.0

Short Window Attention Enables Long-Term Memorization

ICLR 2026Poster

4.4

Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoning

ICLR 2026Rejected

二作

4.0

SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution

NeurIPS 2025Poster

6.3

TestGenEval: A Real World Unit Test Generation and Test Completion Benchmark

ICLR 2025Poster

二作

5.0

SWE-bench Multimodal: Do AI Systems Generalize to Visual Software Domains?

ICLR 2025Poster

4.5

RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning

ICLR 2025Rejected

通讯

4.0

Optimizing Language Models for Inference Time Objectives using Reinforcement Learning

合作者 (20)

Gabriel Synnaeve

Short Window Attention Enables Long-Term Memorization

Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoning

BigO(Bench) - Can LLMs Generate Code with Controlled Time and Space Complexity?

Stochastic activations

What Makes Large Language Models Reason in (Multi-Turn) Code Generation?

The KoLMogorov Test: Compression by Code Generation

RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning

SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution

TestGenEval: A Real World Unit Test Generation and Test Completion Benchmark

SWE-bench Multimodal: Do AI Systems Generalize to Visual Software Domains?

RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning

Optimizing Language Models for Inference Time Objectives using Reinforcement Learning