Leaderboards, eval harnesses and ablations — the contested business of deciding which model is actually better.
arXiv:2610.08825v1 Announce Type: new Abstract: Autonomous driving has made remarkable progress, with recent AI advances enabling commercial deployments that are reshaping urban mobility. Yet the fie...
The paper introduces Mode‑Support Semi‑Discrete Optimal Transport (MoSDOT) to address a failure mode in offline multi‑agent reinforcement learning where flow‑based teachers produce conflicting coordination modes. MoSDOT aggregates multimodal replay into a finite mode support with specified capacities and assigns each noise sample to a single mode before teacher training, thereby preventing the teacher from generating samples between valid modes. Experiments on controlled diagnostics and offline MARL benchmarks show that MoSDOT improves endpoint quality and routing consistency, especially on datasets with multimodal joint behavior.
arXiv:2510.01335v2 Announce Type: replace Abstract: The manifold hypothesis suggests that data lies on manifolds with smaller intrinsic dimension (ID) than their ambient dimension. However there is n...
arXiv:2610.09633v1 Announce Type: new Abstract: The evaluation of coding agents generally strives to be as realistic as possible. In our study, we collect 4,782 agent sessions of real software engine...
Video Prediction Policy 2 (VPP2) is a new world action model that improves zero‑shot generalization for both video prediction and action generation. It achieves this by pretraining a large, diverse manipulation video dataset with event‑level supervision, then distilling the model into a single‑step visual planner and adding a mixture‑of‑transformers action module. Experiments show VPP2 outperforms leading baselines on open‑ended video prediction, real‑world zero‑shot manipulation, and several challenging benchmarks.
The paper investigates how preprocessing choices influence time‑series forecasting performance. By evaluating 11 models across 16 reversible preprocessing pipelines on 29,000 M4 series, it shows that optimizing preprocessing can improve accuracy by 27% to 87%, especially for models without built‑in preprocessing. This demonstrates that simpler architectures can rival complex state‑of‑the‑art models when preprocessing is properly handled.
Multivariate long-horizon forecasting is critical to electricity load scheduling and traffic flow management, and to financial risk control. Existing deterministic backbones output a single trajectory...
LLM agents increasingly improve at inference time by accumulating experience in context rather than by updating parameters. This process is often described as in-context reinforcement learning (ICRL)....
Video anomaly detection (VAD) is a fundamental and safety-critical task in computer vision. Recent generative approaches detect anomalies from a distributional perspective, but remain limited by local...
Generating high-quality and controllable camera motion is essential for AI-assisted cinematography, video synthesis, and 3D scene understanding. We introduce TKCAM, a Text- and Keyframe-conditioned CA...
Neural operators enable fast PDE forecasting, but repeated predictions accumulate errors and fine-scale structures remain difficult to resolve. We introduce a frequency-decomposed finite-time flow-map...
We present Fluid-Gen-Zero, a training-free framework for physics-aware fluid-object interaction video generation that decouples physical reasoning from appearance synthesis. Our key insight is to dele...
Claude Haiku 5.5 is Anthropic’s new fast, low‑cost model that matches GPT‑6 Luna’s pricing of $0.10 per 1,000 input tokens and $0.50 per 1,000 output tokens up to 100,000 tokens, with a 5× increase beyond that threshold. It uses a less generous tokenizer, meaning the same long prompt consumes about 1.25× more tokens than Haiku 4.5, which raises hidden costs. Benchmarks show Haiku 5.5 outperforms Luna within the 100,000‑token limit, but Luna becomes more economical for larger workloads.
Speech quality predictors are increasingly used as rewards, yet no agreed measure of their hackability exists. The usual measurement has two flaws. First, the perturbation reaches the predictor throug...
Video-language models commonly assume that more temporal observations lead to more reliable reasoning. We question this assumption and argue that the key challenge is not merely processing more video...
Large Language Models (LLMs) achieve strong performance across many domains, but their efficiency is limited by the quadratic cost of attention with respect to prompt length. Sparse attention reduces...
Language-model agents are increasingly asked to carry out open-ended scientific research, yet their results are usually graded against a known answer, a rubric, or a language-model reviewer, none of w...
Robot foundation models provide strong visuomotor control, yet their performance can degrade when object positions or task instructions change. Further improvements often require post-training on subs...
Bioacoustics and ethology encompass a wide range of audio understanding tasks, many of which stand to benefit from recent advances in large audio-language models. However, progress in the field has so...
Embodied and assistive agents must do more than recognize objects: they must reason about where an object belongs given the layout of an environment and the habits of the people who live in it. Progre...