works on

From the 1 of 16 linked papers with an AI index.

activity
20242026
collaborators

16 papers

cs.AI2026

Visual Access Boundaries in Vision-Language Model Reasoning

Hiroto Osaka, Shohei Taniguchi, Gouki Minegishi +3

The paper investigates how chain-of-thought prompting works in vision-language models by introducing a visual access sweep that masks attention to image tokens, defining a visual a…

cs.LG2026

On Advantage Estimates for Max@K Policy Gradients

Shota Takashiro, Soichiro Nishimori, Paavo Parmas +6

Reinforcement learning with verifiable rewards is widely used for post-training reasoning models, but sparse outcome rewards make exploration difficult. A complementary approach is…

cs.CL2026

Steering at the Source: Style Modulation Heads for Robust Persona Control

Yoshihiro Izawa, Gouki Minegishi, Koshi Eguchi +2

Activation steering offers a computationally efficient mechanism for controlling Large Language Models (LLMs) without fine-tuning. While effectively controlling target traits (e.g.…

cs.AI2026

Zipping the Thought: When and How Compressed Reasoning Data Works in LLM Post-Training

Kohsei Matsutani, Gouki Minegishi, Takeshi Kojima +2

Large language models (LLMs) can now solve complex problems through long chain-of-thought (CoT) reasoning, but the trade-off between performance and token cost remains a central ch…

cs.AI2026

Emergent Analogical Reasoning in Transformers

Gouki Minegishi, Jingyuan Feng, Hiroki Furuta +3

Analogy is a central faculty of human intelligence, enabling abstract patterns discovered in one domain to be applied to another. Despite its central role in cognition, the mechani…

cs.AI2026

RL Squeezes, SFT Expands: A Comparative Study of Reasoning LLMs

Kohsei Matsutani, Shota Takashiro, Gouki Minegishi +3

Large language models (LLMs) are typically trained by reinforcement learning (RL) with verifiable rewards (RLVR) and supervised fine-tuning (SFT) on reasoning traces to improve the…