1.1k citations · 1.1k across the 25 of their papers we have counts for
29 papers
Who Is the Influencer? Consensus Dynamics in Multi-Agent Systems
Gouki Minegishi, Jingyuan Feng, Hiroki Furuta +3
Multi-agent LLM systems are increasingly used in practice, but their behavior is complex and hard to understand. We design a simple task to ask which model has the most influence o…
Matching Object or Relation? Tracing Abstract Reasoning Inside VLMs
Gouki Minegishi, Hiroki Furuta, Takeshi Kojima +2
Vision Language Models (VLMs) excel on visual benchmarks but fail systematically on tasks requiring abstract reasoning. Existing benchmarks document this failure but cannot say \em…
Few-Shot Demonstrations Elicit the Use of In-Context World Representations in LLMs
Kohsei Matsutani, Gouki Minegishi, Core Francisco Park +3
Large language models (LLMs), when acting as agents, are expected to take observed data in context, infer the latent state space underlying the world, and leverage it for downstrea…
Batch-wise Adaptive Pruning: Periodic Neuron Activation-Aware Weight Pruning for Language Reasoning Model
Yongmin Kim, Shota Takashiro, Yusuke Iwasawa +2
Large Reasoning Models (LRMs) achieve strong performance on complex tasks through extended chain-of-thought generation, but incur substantial computational costs during inference.…
Bootstrapping Niche Multilingual Code Translation via Reinforcement Learning with Execution-Based Verifiable Supervision
Kouki Yuki, Jie Zeng, Kyoko Ogawa +6
Code translation must preserve executable behavior across many programming languages, yet neural code translation has largely focused on a few popular languages such as C++, Java,…
CHASE: Cache-Hole-Adapted Skip Exit for Looped State-Space Language Models
Zhenxuan Yu, Takeshi Kojima, Yutaka Matsuo +1
Recent work on looped language models suggests that many reasoning problems benefit from greater computational depth rather than from additional independent parameters. Existing st…
OrderGrad: Optimizing Beyond the Mean with Order-Statistic Policy Gradient Estimation
Paavo Parmas, Yongmin Kim, Kohsei Matsutani +5
Policy-gradient methods usually optimize expected return, but many real world applications care about distributional properties of returns: tail risk, outlier robustness, or best-o…
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…
Clustered Self-Assessment: A Simple yet Effective Method for Uncertainty Quantification in Large Language Models
Qi Cao, Takeshi Kojima, Andrew Gambardella +3
Large language models (LLMs) demonstrate remarkable performance across diverse tasks, but they often generate responses that appear plausible while being factually incorrect. This…
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…
Understanding Emergent Misalignment via Feature Superposition Geometry
Gouki Minegishi, Hiroki Furuta, Takeshi Kojima +2
Emergent misalignment, where fine-tuning on narrow, non-harmful tasks induces harmful behaviors, poses a key challenge for AI safety in LLMs. Despite growing empirical evidence, it…
Semantic Token Clustering for Efficient Uncertainty Quantification in Large Language Models
Qi Cao, Andrew Gambardella, Takeshi Kojima +2
Large language models (LLMs) have demonstrated remarkable capabilities across diverse tasks. However, the truthfulness of their outputs is not guaranteed, and their tendency toward…
Safe Transformer: An Explicit Safety Bit For Interpretable And Controllable Alignment
Jingyuan Feng, Andrew Gambardella, Gouki Minegishi +3
Current safety alignment methods encode safe behavior implicitly within model parameters, creating a fundamental opacity: we cannot easily inspect why a model refuses a request, no…
ClinDet-Bench: Beyond Abstention, Evaluating Judgment Determinability of LLMs in Clinical Decision-Making
Yusuke Watanabe, Yohei Kobashi, Takeshi Kojima +3
Clinical decisions are often required under incomplete information. Clinical experts must identify whether available information is sufficient for judgment, as both premature concl…
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…
-MoE: Generalizing Mixture of Experts to Infinite Experts
Shota Takashiro, Takeshi Kojima, Shohei Taniguchi +2
The Mixture of Experts (MoE) selects a few feed-forward networks (FFNs) per token, achieving an effective trade-off between computational cost and performance. In conventional MoE,…
Automated Refinement of Essay Scoring Rubrics for Language Models via Reflect-and-Revise
Keno Harada, Lui Yoshida, Takeshi Kojima +2
The performance of Large Language Models (LLMs) is highly sensitive to the prompts they are given. Drawing inspiration from the field of prompt optimization, this study investigate…
When Instructions Multiply: Measuring and Estimating LLM Capabilities of Multiple Instructions Following
Keno Harada, Yudai Yamazaki, Masachika Taniguchi +4
As large language models (LLMs) are increasingly applied to real-world scenarios, it becomes crucial to understand their ability to follow multiple instructions simultaneously. To…
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…
Dynamic Injection of Entity Knowledge into Dense Retrievers
Ikuya Yamada, Ryokan Ri, Takeshi Kojima +2
Dense retrievers often struggle with queries involving less-frequent entities due to their limited entity knowledge. We propose the Knowledgeable Passage Retriever (KPR), a BERT-ba…
Topology of Reasoning: Understanding Large Reasoning Models through Reasoning Graph Properties
Gouki Minegishi, Hiroki Furuta, Takeshi Kojima +2
Recent large-scale reasoning models have achieved state-of-the-art performance on challenging mathematical benchmarks, yet the internal mechanisms underlying their success remain p…
A Comprehensive Survey on Physical Risk Control in the Era of Foundation Model-enabled Robotics
Takeshi Kojima, Yaonan Zhu, Yusuke Iwasawa +8
Recent Foundation Model-enabled robotics (FMRs) display greatly improved general-purpose skills, enabling more adaptable automation than conventional robotics. Their ability to han…
Inconsistent Tokenizations Cause Language Models to be Perplexed by Japanese Grammar
Andrew Gambardella, Takeshi Kojima, Yusuke Iwasawa +1
Typical methods for evaluating the performance of language models evaluate their ability to answer questions accurately. These evaluation metrics are acceptable for determining the…
Which Programming Language and What Features at Pre-training Stage Affect Downstream Logical Inference Performance?
Fumiya Uchiyama, Takeshi Kojima, Andrew Gambardella +3
Recent large language models (LLMs) have demonstrated remarkable generalization abilities in mathematics and logical reasoning tasks. Prior research indicates that LLMs pre-trained…