Communities
Connect sessions
AI calendar
Organizations
Join Slack
Contact Sales

Terms and Conditions

Twitter GitHub LinkedIn Bluesky Youtube

© 2026 ResearchTrend.AI, All rights reserved.

Home
Papers
2507.10616
Cited By

Scalpel vs. Hammer: GRPO Amplifies Existing Capabilities, SFT Replaces Them

v1v2 (latest)

Scalpel vs. Hammer: GRPO Amplifies Existing Capabilities, SFT Replaces Them

13 July 2025

Aryo Pradipta Gema

Seraphina Goldfarb-Tarrant

ArXiv (abs)PDF HTML HuggingFace (1 upvotes)

Papers citing "Scalpel vs. Hammer: GRPO Amplifies Existing Capabilities, SFT Replaces Them"

4 / 4 papers shown

Zooming into Comics: Region-Aware RL Improves Fine-Grained Comic Understanding in Vision-Language Models

Zooming into Comics: Region-Aware RL Improves Fine-Grained Comic Understanding in Vision-Language Models

Sabine Süsstrunk

102

0

0

09 Nov 2025

Towards a Unified View of Large Language Model Post-Training

Towards a Unified View of Large Language Model Post-Training

...

108

11

0

04 Sep 2025

AMFT: Aligning LLM Reasoners by Meta-Learning the Optimal Imitation-Exploration Balance

AMFT: Aligning LLM Reasoners by Meta-Learning the Optimal Imitation-Exploration Balance

235

3

0

09 Aug 2025

Revisiting LLM Reasoning via Information Bottleneck

Revisiting LLM Reasoning via Information Bottleneck

169

10

0

24 Jul 2025