From the 1 of 6 linked papers with an AI index.
6 papers
Budget-Aware LLM Discovery via Cost-Calibrated Frontier Utility
Yansen Zhang, Yilu Liu, Tianyu Liu +6
The paper proposes CostAda, a cost‑aware controller that guides large language model‑based discovery by evaluating frontier progress relative to the token cost incurred, enabling m…
BabyVision: Visual Reasoning Beyond Language
Liang Chen, Weichu Xie, Yiyan Liang +27
While humans develop core visual skills long before acquiring language, contemporary Multimodal LLMs (MLLMs) still rely heavily on linguistic priors to compensate for their fragile…
G1: Bootstrapping Perception and Reasoning Abilities of Vision-Language Model via Reinforcement Learning
Liang Chen, Hongcheng Gao, Tianyu Liu +5
Vision-Language Models (VLMs) excel in many direct multimodal tasks but struggle to translate this prowess into effective decision-making within interactive, visually rich environm…
PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling
Zefan Cai, Yichi Zhang, Bofei Gao +8
In this study, we investigate whether attention-based information flow inside large language models (LLMs) is aggregated through noticeable patterns for long context processing. Ou…
MMEvalPro: Calibrating Multimodal Benchmarks Towards Trustworthy and Efficient Evaluation
Jinsheng Huang, Liang Chen, Taian Guo +13
Large Multimodal Models (LMMs) exhibit impressive cross-modal understanding and reasoning abilities, often assessed through multiple-choice questions (MCQs) that include an image,…
Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models
Bofei Gao, Feifan Song, Zhe Yang +17
Recent advancements in large language models (LLMs) have led to significant breakthroughs in mathematical reasoning capabilities. However, existing benchmarks like GSM8K or MATH ar…