arXiv:2604.09870v1 Announce Type: new Abstract: We investigate how looped transformers encode human preference in their internal iteration states. Using Ouro-2.6B-Thinking, a 2.6B-parameter looped transformer with iterative refinement, we extract hidden states from each loop iteration and train lightweight evaluator heads…
arXiv:2503.16771v3 Announce Type: replace-cross Abstract: As Large Language Models for Code (LM4Code) become integral to software engineering, establishing trust in their output becomes critical. However, standard accuracy metrics obscure the underlying reasoning of generative models, offering little insight into how…
arXiv:2604.09876v1 Announce Type: new Abstract: Generative user interfaces (UIs) create new opportunities to adapt interfaces to individual users on demand, but personalization remains difficult because desirable UI properties are subjective, hard to articulate, and costly to infer from sparse feedback.…
arXiv:2510.27269v3 Announce Type: replace-cross Abstract: Reasoning language models (RLMs) achieve strong performance on complex reasoning tasks, yet they still exhibit a multilingual reasoning gap, performing better in high-resource languages than in low-resource ones. While recent efforts have been made to…
arXiv:2604.09887v1 Announce Type: new Abstract: Medical vision-language datasets are often limited in size and biased toward negative findings, as clinicians report abnormalities mostly but might omit some positive/neutral findings because they might be considered as irrelevant to the patient’s condition.…
arXiv:2602.22822v2 Announce Type: replace-cross Abstract: The identification and property prediction of chemical molecules is of central importance in the advancement of drug discovery and material science, where the tandem mass spectrometry technology gives valuable fragmentation cues in the form of…
arXiv:2604.09905v1 Announce Type: new Abstract: Emergency department triage relies heavily on both quantitative vital signs and qualitative clinical notes, yet multimodal machine learning models predicting triage acuity often suffer from modality collapse by over-relying on structured tabular data. This limitation…
arXiv:2604.11328v1 Announce Type: cross Abstract: Automatic prompt optimization (APO) hinges on the quality of its evaluation signal, yet scoring every prompt candidate on the full training set is prohibitively expensive. Existing methods either fix a single evaluation subset before optimization…
arXiv:2604.09909v1 Announce Type: new Abstract: We study last-iterate convergence of SGD with greedy step size over smooth quadratics in the interpolation regime, a setting which captures the classical Randomized Kaczmarz algorithm as well as other popular iterative linear system solvers.…
arXiv:2604.11611v1 Announce Type: cross Abstract: To overcome the sparse reward challenge in reinforcement learning (RL) for agents based on large language models (LLMs), we propose Mutual Information Self-Evaluation (MISE), an RL paradigm that utilizes hindsight generative self-evaluation as dense reward…
