Research on medical item difficulty has largely focused on prediction rather than on identifying interpretable patterns associated with empirical difficulty. This study used empirical difficulty labels to guide large language models (LLMs) in inducing difficulty rubrics and examined how rubric guidance affected LLM dif...
Xiaomeng Xiong, Marcus Walker, Yanlin Jiang· 0 citations
The decoding order of masked diffusion language models (dLLMs) can substantially affect generation quality, yet common decoders typically apply the same reveal rule throughout generation. Confidence provides a strong default, but it measures how decisively a position can be predicted now, not how useful revealing that...
Chengwei Zhou, Zhaoyan Jia, Massoud Pedram et al.· 0 citations
Generalized Category Discovery (GCD) aims to perform old-category recognition and novel-category discovery in the same unlabeled set. Early methods address both objectives through clustering and evaluate the exchangeable predicted indices with global Hungarian (GH) matching. Recent methods adopt trainable parametric he...
Zitai Wang, Qianqian Xu, Cong Hua et al.· 0 citations
Frontier language models such as GPT 6 Astra have contributed to research mathematics, demonstrating substantial reasoning capabilities. Yet serious reasoning failures persist even in tasks governed by explicit rules. We demonstrate this in retrograde chess which is a playground for pure reasoning, where solvers recons...
Tidor-Vlad Pricope· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Vision-language models (VLMs) are increasingly prompted with personas, such as "You have low vision", to stand in for people in user studies. We ask whether a persona makes a model see like the person it describes. We compare persona answers with measured people with low vision, color blindness or hearing loss, and wit...
This work introduces Learning to Bound (L2B), a data-driven framework that amortizes bound computation cost by training a neural predictor and targets two key properties required for effective search bounds: admissibility and progressive tightening as evidence accumulates.
Akshay Vyas, Hai-Liang Dong, Shivvrat Arya et al.· 0 citations
It is proved that under exclusive specialization every update that raises pass@1 strictly lowers the success of some prompt; a strictly positive price on prompts (Stiemke duality) is the exact condition, and being merely the best strategy on a prompt is provably not enough.
Predicting future physical states stands as a central challenge in embodied intelligence. Existing video world models leave 3D dynamics implicit, action-based formulations remain tied to specific robot configurations, and 3D motion prediction models use flow-only supervision, providing a limited learning signal for phy...
Yu-Hao Lu, Jiong-Xu Chen, He-Xiang Wei et al.· 0 citations
Multi-image composition (MICo) is a critical yet challenging capability for large-scale image editing models, where consistency degrades rapidly as the number of reference images increases. While supervised fine-tuning (SFT) can equip models with basic MICo ability, it provides only implicit supervision on the composed...
Zhiqiang Wu, Shuang Sun, Jiale Zhang et al.· 0 citations
This work proposes profiling-guided layer-wise RoPE allocation, assigning high/low coverage under a fixed budget using early coverage-benefit profiles, and normalized-depth mapping transfers profiles within model families, enabling small-model measurements to guide larger-model allocations before pretraining.
Experience can help embodied agents turn high-level plans into grounded actions, but transferring it across tasks remains challenging: previously effective strategies may be inapplicable to the current state or fail to produce their intended outcomes. We present EmbodiExp, a framework for constructing transferable expe...
Zhifeng Gu, Ying Chen, Rui Jiang et al.· 0 citations
Sequential recommenders typically construct history inputs from item identifiers or content features using an embedding map shared across users. The same item therefore yields the same initial embedding representation, even when users have different relationships to the item. We propose personalized item representation...
Renzhi Wu, Zi-Kun Cui, Jun-Jie Yang et al.· 0 citations
