Leaderboards, eval harnesses and ablations — the contested business of deciding which model is actually better.
arXiv:2509.25692v2 Announce Type: replace-cross Abstract: Active Test-Time Adaptation (ATTA) improves model robustness under domain shift by selectively querying human annotations at deployment, but...
arXiv:2601.11433v2 Announce Type: replace Abstract: Deep Differentiable Logic Gate Networks (LGNs) and Lookup Table Networks (LUTNs) are demonstrated to be suitable for the automatic classification o...
The paper introduces Global Relative Kinetic Utility (Global RKU), a label‑free method for calibrating cross‑layer credit in global structured pruning of large language models. Global RKU estimates channel importance via a final‑hidden‑state activation‑gradient signal and applies block‑relative normalization to remove block‑common scale while preserving within‑block ordering, enabling a single‑stage static pruning topology. Experiments on Qwen‑2.5‑7B show significant performance gains at various sparsity levels, and ablation studies confirm the effectiveness of the relative‑normalization step.
arXiv:2608.26086v3 Announce Type: replace-cross Abstract: Auto-research agents now run machine-learning development unattended for hours, revising data pipelines, models, and validation from their ow...
arXiv:2609.34426v2 Announce Type: replace Abstract: This paper addresses the problem of safe offline reinforcement learning, which involves training a policy to satisfy safety constraints using an of...
The paper introduces Alignment‑Guided Flow Transformer (AGFT), a framework for Vision‑Language‑Action (VLA) models that explicitly enforces tri‑modal alignment among vision, language, and action through a dedicated alignment loss. AGFT bridges representational gaps across modalities, improving task adaptation and robustness, and employs a flow‑matching objective to reduce inference steps compared to diffusion‑based policies. Experiments on a large benchmark demonstrate that AGFT achieves higher success rates and lower inference latency than state‑of‑the‑art baselines, highlighting tri‑modal alignment as crucial for scalable VLA manipulation.
ALICE is a foundation model that estimates mutual information (MI) without per‑distribution training. Trained only on synthetic distributions, it acts as an in‑context estimator of rectified‑flow velocity fields, producing MI via a fixed identity that integrates squared differences between joint and conditional fields. The authors validate ALICE on a challenging benchmark and demonstrate its applicability to unseen data in biology, genetics, and neuroscience, achieving performance comparable to neural estimators trained separately for each distribution.
arXiv:2410.14839v5 Announce Type: replace-cross Abstract: We study the dynamic pricing problem faced by a broker seeking to learn prices for a large number of credit market securities, such as corpor...
arXiv:2501.12632v3 Announce Type: replace-cross Abstract: Weakly supervised object localization (WSOL) models can predict both the object class and the spatial regions corresponding to the object, wi...
arXiv:2505.05261v4 Announce Type: replace-cross Abstract: Two-stage stochastic programming (2SP) offers a basic framework for modelling decision-making under uncertainty, yet scalability remains a ch...
The paper introduces AttWarp, a lightweight technique that uses a multimodal large language model’s cross‑modal attention to perform rectilinear warping of input images at test time. By reallocating spatial resolution toward query‑relevant regions without altering model weights or architecture, AttWarp preserves global context while making small objects and subtle relationships easier for the model to read. Experiments on five benchmarks and four MLLMs show consistent accuracy gains, improved compositional reasoning, and reduced hallucinations compared to baseline image‑manipulation methods.
arXiv:2603.28773v2 Announce Type: replace-cross Abstract: Large language models (LLMs) frequently generate confident yet factually incorrect content when used for language generation (a phenomenon of...
arXiv:2604.21863v2 Announce Type: replace-cross Abstract: Deep reinforcement learning for quantum circuit optimization faces three bottlenecks: replay buffers that overlook temporal difference (TD) t...
The paper introduces Minimum Specification Perturbation (MSP), a metric that counts the smallest number of analyst decisions that must be altered to make a causal study’s confidence interval include zero. MSP is small under the null hypothesis, grows with effect size, and provides a distance‑to‑falsification measure that traditional dispersion‑based robustness tools cannot capture. The authors demonstrate that MSP and the Fragility Index assess different vulnerabilities, and show that on the LaLonde benchmark MSP equals one, meaning a single decision change would render the estimate statistically insignificant.
arXiv:2605.11608v2 Announce Type: replace-cross Abstract: A single base LLM now comes with dozens of post-training variants, quantized, LoRA-adapted, or distilled, and each has to be checked before r...
arXiv:2605.29139v2 Announce Type: replace-cross Abstract: Question-answering services built on retrieval-augmented generation (RAG), in which a language model answers from retrieved documents, are in...
arXiv:2606.16996v2 Announce Type: replace-cross Abstract: Segment Anything Model 3 (SAM 3) provides a strong frozen backbone for concept-prompted segmentation, but applying it directly to open-vocabu...
arXiv:2609.33180v2 Announce Type: replace-cross Abstract: As recursive self-improvement (RSI) rapidly advances, reliable evaluation becomes critical for guiding adaptive search. RSI typically relies...
The paper introduces Event‑Grounded Self‑Distillation (EGSD), a method for real‑time video understanding that treats streaming memory as incremental updates over verifiable events such as key visual entities, actions, and details. EGSD adapts on‑policy self‑distillation by combining a multiplicative weight with outcome reward to align teacher and student preferences, and re‑weights the teacher with event information plus an entity‑coverage reward to counter question‑relevance bias. Experiments on StreamingBench and OVO‑Bench Real‑Time track show EGSD achieves 79.8 % and 73.4 % accuracy respectively, while improving effective‑entity recall by 17.4 % with only a 6.8 % increase in memory length.
arXiv:2609.37098v1 Announce Type: cross Abstract: Vehicle-infrastructure cooperation can complement onboard sensing with broader and more informative observations of the traffic environment, providin...