Leaderboards, eval harnesses and ablations — the contested business of deciding which model is actually better.
arXiv:2609.38604v1 Announce Type: cross Abstract: Modern LLM agents increasingly tackle complex tasks through interactive, long-horizon exchanges with users, while existing benchmarks generally assum...
arXiv:2609.38653v1 Announce Type: cross Abstract: Recent advances in musculoskeletal modeling and reinforcement learning have enabled muscle-actuated agents to reproduce increasingly complex human mo...
arXiv:2609.38811v1 Announce Type: new Abstract: Metal additive manufacturing parts are inspected by X-ray computed tomography, where labelled data is scarce, the pores and inclusions that matter span...
The paper introduces a causal decomposition framework to diagnose failures in multimodal large language models (MLLMs) by distinguishing between intrinsic capability deficits and cascading errors from upstream prerequisites. It defines capability metrics (NC, IC, RC) and contribution metrics (N-Score, S-Score) to evaluate tasks under different prerequisite conditions, and applies the framework to the CADET benchmark of 10 composite tasks with 46 unit tasks and 33,000 human-annotated questions. The study finds that providing correct prerequisites eliminates 54% of errors on cognitive tasks and that a single critical prerequisite can account for 84% of the improvement gained by supplying all prerequisites.
CellMSA introduces a novel single‑cell representation learning framework that leverages a multiple‑sequence‑alignment‑inspired context model. For each target cell, it retrieves relevant cells across batches and related cell types, summarizing cross‑cell patterns into a context‑dependent gene‑pair representation that is fed into a pair‑aware encoder. Pretraining on a massive human single‑cell corpus (≈109 million cells) and subsequent benchmarks demonstrate consistent performance gains over existing methods.
arXiv:2609.38914v1 Announce Type: new Abstract: Evaluating interactive agents is expensive. Agent behavior is stochastic, so reliability must be measured over repeated trials, but failures are rare a...
arXiv:2609.38925v1 Announce Type: new Abstract: Multimodal federated learning (MFL) has emerged as a pivotal paradigm for leveraging distributed data to enhance model performance. However, existing m...
arXiv:2609.39229v1 Announce Type: cross Abstract: Automatic evaluation of faithfulness increasingly relies on a large language model acting as a judge, yet the most reliable judges are proprietary fr...
arXiv:2609.39367v1 Announce Type: new Abstract: Despite the widespread use of Low-Rank Adaptation (LoRA), little is known about its dynamics in continual learning and the mechanisms by which low-rank...
The paper introduces FAME, a training‑free framework that evaluates false memory in autonomous agents by tracking how their internal beliefs shift under counterfactual scenarios. False memory, defined as biases arising from spurious correlations, environment shifts, or knowledge conflicts, is hard to detect with standard methods. FAME measures concept drift in hidden states, achieving AUROCs between 76.2% and 96.7% and outperforming baselines on benchmarks such as GSM‑Symbolic, GitChameleon, and BigBench‑Hard.
arXiv:2609.39588v1 Announce Type: new Abstract: We push the frontier of large-scale spatial intelligence in Vision-Language Models (VLMs) and introduce the first benchmark that probes geographical la...
arXiv:2609.39645v1 Announce Type: cross Abstract: Multi-agent collaboration lets large language models (LLMs) improve question answering through deliberation and feedback. Yet shared discussion coupl...
arXiv:2609.39687v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) trains mathematical reasoning models using a privileged teacher that sees a reference solution and supervises studen...
arXiv:2609.39701v1 Announce Type: new Abstract: Value steering should change an LLM's normative priorities while preserving the scenario, facts, and task constraints underlying its answer. Convention...
arXiv:2609.39836v1 Announce Type: new Abstract: Contrastive vision-language models map visual and textual representations into a shared normalized embedding space, making cosine similarity the natura...
BayesNDE is a neural density estimator that uses Bayesian generative modeling to estimate densities without relying on invertible networks or Jacobian-determinant calculations. It constructs an adaptive proposal for each observation by inferring a sample-specific latent posterior, and then applies bridge sampling to combine proposal samples with separate posterior samples for density estimation. Experiments on synthetic datasets show improved density estimation and structure recovery, while real-world applications demonstrate better anomaly detection.
arXiv:2609.39884v1 Announce Type: cross Abstract: Role prompting elicits specialized behavior from large language models through an expert identity, offering a lightweight way to guide reasoning on d...
The paper introduces the Cluster Attention Neural Operator (CANO), a neural operator that uses a novel cross‑attention mechanism to dynamically cluster queries while keeping full‑resolution keys and values. This design eliminates the slice compression and weight‑sharing limitations of previous Transformer‑based operators, maintaining fast computation and global interactions. Experiments on a range of fluid and solid dynamics benchmarks—including Navier‑Stokes, Airfoil, Plasticity, Pipe Turbulence, and Composites—show that CANO achieves lower errors than existing baselines and demonstrates strong geometric adaptability and temporal consistency.
arXiv:2609.40305v1 Announce Type: new Abstract: Improving text-to-image models has traditionally relied on increasing model size or the number of denoising steps. In this work, we explore an alternat...
NeuroAtlas is the largest EEG benchmark to date, comprising 42 datasets and 260,000 hours of clinical EEG data across epilepsy, sleep medicine, brain age estimation, and brain‑computer interfaces. The study evaluates foundation models (FMs) for EEG against supervised baselines and generic time‑series FMs, finding that EEG‑specific FMs do not consistently outperform generic ones. It also demonstrates that standard machine‑learning metrics are inadequate for clinical relevance, advocating for task‑specific measures such as event‑level decision quality, hypnogram features, and brain‑age gap.