Repository navigation
Paper Review: GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning - #133
Merged
Merged
Conversation
…perform Reinforcement Learning'
- TL;DR: 9.2x shorter is Qwen3 8B PUPA (Fig. 18b); on GPT-4.1 Mini the maximum is 7.5x (HotpotQA, Fig. 18a). - Merge: the paper caps it at 5 calls (App. E.4); it never says lineage conditions make it rare. - The five GRPO margins are unnamed in the paper; the post says they match the Table 1 differences, in percentage points. - IFBench curve sentence read off the figure is replaced by the caption's claim. - AIME-2025 +12% matches the Qwen3 8B table (32.00 vs 20.00), not GPT-4.1 Mini.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR adds an AI-generated paper review: GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
This pull request was automatically generated by the Scholar-Lens system.
Review checklist
aie::vector사용법 같은 아키텍처 지식이 지시문과 템플릿 형태로 담겨 있습니다." — 원문 본문에 없고 그림 분석(이미지 판독)에서만 읽어 낸 내용입니다. 그림에서 값을 읽Hand fixes
Checked, unchanged: AIME-2025 / LiveBench-Math feedback cells already read "원문에 명시 없음" (App. E.1 describes only four); A/B Pareto table is the post's labelled illustration; AIME 2022–2024 train/val and AIME-2025 test (App. E.1); all Table 1/2/3 values.