9 papers
Task-CoEvolve: Efficient Harness Optimization via Adaptive Validation Task Selection
Atsuyuki Miyai, Kiyoharu Aizawa, Toshihiko Yamasaki
We present a novel approach to efficient LLM harness optimization through adaptive validation task selection. Harness optimization iteratively rewrites the harness code based on va…
360CityArena: A Realistic Virtual Urban Navigation Benchmark for Embodied Agents
Kenta Watanabe, Atsuyuki Miyai, Mizuki Takenawa +2
We present 360CityArena, a benchmark for evaluating the urban exploration capabilities of embodied agents within a photorealistic environment constructed from 360-degree videos. Ex…
Beyond Bilingual Transfer: Multilingual Code-Switching in Instruction Tuning
Shunta Asano, Jeonghun Baek, Toshihiko Yamasaki
Recent studies have shown that code-switching data (CSD), in which multiple languages are mixed within the same context, can improve cross-lingual transfer and multilingual alignme…
Paper Reconstruction Evaluation: Evaluating Presentation and Hallucination in AI-written Papers
Atsuyuki Miyai, Mashiro Toyooka, Zaiying Zhao +3
This paper introduces the first systematic evaluation framework for quantifying the quality and risks of papers written by modern coding agents. While AI-driven paper writing has b…
Mirai: Autoregressive Visual Generation Needs Foresight
Yonghao Yu, Lang Huang, Zerun Wang +2
Autoregressive (AR) visual generators model images as sequences of discrete tokens and are trained with a next-token likelihood objective. This strict causal supervision optimizes…
CE-FAM: Concept-Based Explanation via Fusion of Activation Maps
Michihiro Kuroki, Toshihiko Yamasaki
Although saliency maps can highlight important regions to explain the reasoning behind image classification in artificial intelligence (AI), the meaning of these regions is left to…
Bias Beyond Demographics: Probing Decision Boundaries in Black-Box LVLMs via Counterfactual VQA
Zaiying Zhao, Toshihiko Yamasaki
Recent advances in large vision-language models (LVLMs) have amplified concerns about fairness, yet existing evaluations remain confined to demographic attributes and often conflat…
MobileCity: An Efficient Framework for Large-Scale Urban Behavior Simulation
Xiaotong Ye, Nicolas Bougie, Toshihiko Yamasaki +1
Generative agents offer promising capabilities for simulating realistic urban behaviors. However, existing methods oversimplify transportation choices, rely heavily on static agent…
Joint Fusion and Encoding: Advancing Multimodal Retrieval from the Ground Up
Lang Huang, Qiyu Wu, Zhongtao Miao +1
Information retrieval is indispensable for today's Internet applications, yet traditional semantic matching techniques often fall short in capturing the fine-grained cross-modal in…