Leaderboards, eval harnesses and ablations — the contested business of deciding which model is actually better.
arXiv:2609.22566v1 Announce Type: cross Abstract: Knowledge distillation (KD) aims to compress high-performance teacher LLMs into lightweight students. However, distilled students often exhibit subst...
arXiv:2609.22582v1 Announce Type: new Abstract: End-to-end and vision-language-action (VLA) driving policies are compared by leaderboard rank, but a rank reports an outcome, not the behaviour behind...
arXiv:2609.22700v1 Announce Type: new Abstract: Step-level reasoning evaluators are commonly based on autoregressive language models, whose causal attention restricts each step representation to the...
arXiv:2609.22771v1 Announce Type: cross Abstract: Recent advances in Audio LLMs have achieved human-level speech recognition, yet existing systems struggle to capture paralinguistic aspects such as s...
MolSC is a new dataset of 181,000 substituent-level examples that captures how attaching specific substituents to molecular scaffolds changes properties such as bioactivity and physicochemical descriptors. The authors also provide MolSC-Bench, a held‑out benchmark of 1,541 examples that are disjoint from MolSC at scaffold, substituent, and molecule levels. Experiments show that training molecular large language models on MolSC markedly improves their ability to predict substituent contributions, outperforming existing models on a range of downstream chemistry tasks.
ValueDiff introduces a value‑geometric KV cache eviction strategy for large language models that suppress attention sinks. It ranks tokens by the L2 deviation of their value vectors from the cache mean, a score that aligns with minimal‑disturbance eviction under a max‑entropy assumption. Across several benchmarks—RULER, LongBench, and MATH‑500—ValueDiff consistently retains a higher proportion of useful tokens than prior methods, especially under tight cache budgets.
RPMem introduces a two‑stage architecture that compiles each session into a model‑independent latent memory and then consolidates it with retained memory via a task‑trained recurrent gate. The consolidated memory is mapped to backbone‑specific low‑rank adaptation (LoRA) parameters, enabling the memory to transfer when the backbone is replaced. Across three long‑term memory benchmarks and five diverse backbones, RPMem achieves broad generalization with near‑constant update cost and memory footprint, outperforming existing parametric and text‑based baselines on the PERMA benchmark.
arXiv:2609.23492v1 Announce Type: new Abstract: Perception for embodied agents is video-based, often multi-view (ego, exo, or both), and inherently continual, with simultaneous task and viewpoint shi...
arXiv:2609.23590v1 Announce Type: new Abstract: Battery energy storage system (BESS) dispatch must preserve operational feasibility while declining price spreads reduce the margin available to pay fo...
arXiv:2609.23596v1 Announce Type: cross Abstract: With face-recognition models now embedded in everyday authentication and surveillance, recent works have pinpointed a critical weakness: these models...
arXiv:2609.23600v1 Announce Type: new Abstract: Prompt learning efficiently adapts vision-language models (VLMs) to downstream tasks, but gains on seen classes often come at the expense of generaliza...
arXiv:2609.23601v1 Announce Type: new Abstract: Long-video understanding must capture transient visual evidence under strict token budgets, yet existing methods compress frames, append memory tokens,...
arXiv:2609.23716v1 Announce Type: cross Abstract: Textual-gradient methods automate prompt optimization through natural-language feedback, but their iterative updates can be unstable. We identify two...
arXiv:2609.23825v1 Announce Type: new Abstract: We present a comprehensive benchmark of Federated Learning (FL) for multilingual Automatic Speech Recognition (ASR), evaluating four Speech-LLM archite...
arXiv:2609.23875v1 Announce Type: new Abstract: The rapid growth of resident space objects is increasing the complexity of space situational awareness sensor tasking, challenging classical optimizati...
arXiv:2609.23980v1 Announce Type: cross Abstract: AI agents now report vulnerabilities faster than maintainers can review them. Reports often depend on security properties specific to the application...
arXiv:2609.24026v1 Announce Type: new Abstract: In this paper, we investigate the limitations of fixed, hand-crafted connectors in hierarchical semantic representations for open-vocabulary object det...
The paper introduces Action‑Slot, a structured action‑centric representation learning framework for multi‑agent atomic activity understanding. It reformulates slot attention into activity‑aligned slots, parallel spatio‑temporal updates, and background regularization to disentangle concurrent, asynchronous activities directly from raw video. Additionally, an attention‑difference pseudo‑mask method enables weakly supervised localization, and a new synthetic dataset, TACO, provides balanced atomic activity coverage with pixel‑level annotations.
arXiv:2609.24137v1 Announce Type: cross Abstract: Traditional data systems face profound limitations in the AI era, relying on human-crafted pipelines, lacking semantic understanding of heterogeneous...
arXiv:2609.24151v1 Announce Type: new Abstract: Four-dimensional (4D) Radar has emerged as a key sensor for environmental perception, providing range, azimuth, elevation, and Doppler measurements whi...