Leaderboards, eval harnesses and ablations — the contested business of deciding which model is actually better.
Despite the success of foundation models in image synthesis, achieving fine-grained spatial control remains a significant challenge, typically requiring auxiliary structural adapters or complex infere...
Graph-level out-of-distribution (OOD) generalization is critical for deploying graph neural networks in real-world applications where distribution shifts are inevitable. Existing approaches typically...
Real-world datasets often follow a long-tailed distribution, making generalization to tail classes difficult. We revisit this problem through the lens of shortcut learning, where models prefer the eas...
Mid-training (MT) is a training phase in LLM training pipelines that continues pretraining (PT) on smaller datasets of higher value. Taking the PT and MT phases as a whole, training usually starts wit...
Heuristic solvers for general-sum extensive-form games produce a stream of iterates, and practice reports the NashGap of the returned profile. We ask what a solver can know about its own iterates from...
The main challenge in applying deep neural networks to real-world scenarios lies in distributional discrepancies between training and test datasets, often caused by variations in data acquisition cond...
Language-model agents are increasingly asked to run experiments, yet benchmarks rarely measure whether those experiments improve the decisions they are meant to inform. We introduce CrashGym, a determ...
Streaming video understanding (SVU) requires video-LLMs to process incoming video continuously while answering user queries with low latency. Recent methods reduce query-time cost by prefilling frames...
Personalized reward models require user-specific preference comparisons, but collecting such feedback is costly and provides only limited supervision for each user. At the same time, large pools of re...
Autonomous industrial inspection requires strictly deterministic real-time inference under tight memory and computational constraints. However, existing Video Instance Segmentation (VIS) models suffer...
Random-Forest wrapper feature selection can use out-of-bag (OOB) predictions as a built-in validation signal, but the criteria that decide each forward step accumulate OOB estimation noise. We propose...
arXiv:2610.27816v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) has become the dominant paradigm for grounding the outputs of large language models (LLM) in external knowledge....
arXiv:2610.27867v1 Announce Type: cross Abstract: Prompt-based contextual ASR requires selecting a compact hotword subset from large dynamic inventories, yet existing pre-selection methods rely eithe...
arXiv:2610.27891v1 Announce Type: cross Abstract: Industrial fluid transport systems are monitored through heterogeneous sensing streams, but robust fault diagnosis remains difficult because response...
arXiv:2610.27901v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly used as research collaborators for hypothesis refinement, methodological planning, and technical analys...
arXiv:2610.28416v1 Announce Type: cross Abstract: Code-based video generation offers fine-grained control, targeted editing, and component reuse compared with direct video generation. This approach r...
arXiv:2610.28419v1 Announce Type: cross Abstract: Vision-language trackers (VLTs) accept descriptions intended to assist tracking, but accepting text does not establish that its content changes track...
EvaOmni is a new evidence‑grounded benchmark for audio‑visual reasoning, comprising 392 videos and 1,027 human‑annotated question‑answer pairs that span 15 capabilities and three difficulty levels. The benchmark requires models to provide both an answer and the supporting audio‑visual evidence intervals, and introduces Coverage Grounded (C.G.) and Bottleneck Grounded (B.G.) scores to evaluate evidence identification and temporal localization. Across 24 Omni LLMs, Gemini3.7‑Flash achieved the highest overall accuracy, yet its B.G. scores reveal significant room for improvement in grounded reasoning.
The paper introduces SelfDx, a benchmark that tests whether AI models appropriately respond to health queries by either cautioning against self-diagnosis or agreeing when evidence rules out serious alternatives. It evaluates 27 frontier models on 100 paired cases, finding that most models inappropriately agree in caution-required scenarios, achieving only a 77.0% appropriate caution rate. The study shows that prompt engineering can raise caution but at the cost of agreement, while supervised fine‑tuning on balanced examples improves both metrics.
MemDev-Bench is a new benchmark that evaluates how memory systems influence code agents across diverse repository histories. It pairs each task with relevance‑annotated earlier tasks from the same repository, creating 4,869 historical tasks for 555 base tasks and an automated pipeline that ensures these tasks are executable and verifiable. The study finds that memory mainly improves fine‑grained localization and editing, but can also cause failures when agents reuse memory without reassessing its applicability, and that providing raw historical trajectories reduces steps more effectively than compressed memories.