Search: “model-releases”

Learning to Act under Visual Interruptions with Vision-Language-Action Models

arXiv:2609.35003v1 Announce Type: new Abstract: Vision-language-action (VLA) models have demonstrated strong capabilities in robotic manipulation, but they are typically developed and evaluated with a

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

Learning to Steer, Steering to See: Unveiling the Geometry of RLVR in Large Language Models via Trainable Vectors

DGX agent

arXiv:2609.34344v1 Announce Type: new Abstract: Reinforcement learning (RL) has become a key paradigm for enhancing the reasoning of large language models, yet the high dimensionality of parameter upd

model-releasesarxiv-cs-lg

29 Sept 2026

Learning When to Recur: Token-Adaptive Recursion for Imbalanced Ophthalmic Domain Incremental Learning

arXiv:2609.32785v1 Announce Type: cross Abstract: Domain incremental learning is essential for adapting ophthalmic deep learning models to sequential clinical domains while preserving diagnostic exper

Leveraging Soft Prompts for Privacy Attacks in Federated Prompt Tuning

arXiv:2601.06641v3 Announce Type: replace Abstract: Membership inference attacks (MIAs) pose a serious privacy threat in federated learning (FL). While MIAs have been extensively studied in standard F

Livin' on a Prior: Likelihood Score Approximation for Inverse Problems

arXiv:2609.34446v1 Announce Type: new Abstract: Generative models have found great success as data-driven methods of solving inverse problems. Two popular approaches work either by combining a pretrai

LLM Unlearning Evaluation with TRIAGE

arXiv:2609.32103v1 Announce Type: new Abstract: Large language models can memorize private or harmful information, motivating machine unlearning methods that remove targeted knowledge while preserving

LLM4Trust: Exploring the Capabilities of Large Language Models for Trust Evaluation

arXiv:2609.33521v1 Announce Type: new Abstract: Trust evaluation plays a critical role in cybersecurity by supporting risk mitigation and decision-making. A variety of trust evaluation methods have be

LLMs are General Asynchronous Agents

arXiv:2609.35427v1 Announce Type: new Abstract: Modern LLMs are increasingly capable as autonomous agents, but they follow sequential interaction cycles: read, think, reply or call tools, repeat. Many

LLMs as Adaptive Meta-Solvers: Strategy-Diverse RL for Industrial-Scale Optimization

arXiv:2609.34427v1 Announce Type: new Abstract: Scaling LLM-based optimization from textbook-scale instances to real-world, industrial tasks remains a critical open challenge. Existing approaches are

LLN: Learnable Lens Networks for Parameter-Efficient Long-Horizon Dynamical Prediction

arXiv:2609.34493v1 Announce Type: new Abstract: Explicit residual connections of the form (x+f(x)), often combined with normalization layers, have become a standard strategy for training very deep neu

LogicEnvGen: Task-Logic Driven Generation of Diverse Simulated Environments for Embodied AI

arXiv:2601.13556v2 Announce Type: replace Abstract: Simulated environments play an essential role in embodied AI, functionally analogous to test cases in software engineering. However, existing enviro

Long Time No See: Benchmarking VLMs for Out-of-Sight Spatiotemporal Reasoning in Egocentric Videos

arXiv:2609.34630v1 Announce Type: new Abstract: Real-world AI systems must reason about objects that are no longer visible: an AR assistant guiding a user back to an object used earlier, a household r

Look Before You Judge: Training-Free Region Mining for Grounded and Explainable Deepfake Detection

arXiv:2609.35536v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) can explain deepfake verdicts in natural language, but such explanations are not necessarily visually grounded

Look Before You Select: Rethinking Vocabulary Sparsification in On-Policy Distillation

arXiv:2609.34386v1 Announce Type: cross Abstract: On-policy distillation (OPD) uses teacher correction on student-generated responses. Full-vocabulary correction can provide important corrections even

Loop Dropout: Regularizing Shared Updates in Looped Language Models

arXiv:2609.34218v1 Announce Type: new Abstract: Looped language models separate computational depth from parameter count by repeatedly applying the same transformer block. Adapting these models requir

LoopTrack: A Simple Baseline for Parameter-Efficient Transformer Tracking

arXiv:2609.33306v1 Announce Type: new Abstract: Current Transformer-based tracking methods typically stack multiple Transformer blocks with separate parameters to model interactions between the target

Low-Rank Single-Index Bandits with Unknown Links: From Matrices to Tensors

arXiv:2609.33025v1 Announce Type: new Abstract: Low-rank matrix and tensor bandits exploit structured interactions but typically assume a known reward link. Recent single-index bandit methods accommod

LRC-JEPA: Disentangling Dynamics and Residual Context for Efficient World Models

arXiv:2609.34375v1 Announce Type: new Abstract: Compact JEPA world models enable efficient latent-space planning, but low-dimensional representation trained under reward-free self-supervision must enc

MAESTRO: a Multimodal Auditory-attention Egocentric Speech-TRacking Open corpus

arXiv:2609.31898v1 Announce Type: cross Abstract: Humans rely on gaze, head movements, and visual cues to attend to speakers in noisy environments, yet auditory attention decoding (AAD) has been studi

Maintaining Benchmarks Against Increasingly Capable Agents: Detection and Remediation of Unearned Passes

arXiv:2609.34262v1 Announce Type: cross Abstract: Agentic benchmarks guide model selection and training. Yet an agent can pass a task without demonstrating the intended capability. Such outcomes const

Manifold limit for the training of shallow graph convolutional neural networks

arXiv:2601.06025v2 Announce Type: replace-cross Abstract: We study the discrete-to-continuum consistency of the training of shallow graph convolutional neural networks (GCNNs) on proximity graphs of s

Many Eyes, One World: Feed-Forward 3D Reconstruction from Mixed Cameras

arXiv:2609.35658v1 Announce Type: new Abstract: Real-world capture is heterogeneous: perspective, fisheye, and 360^irc panoramic images can coexist within a single reconstruction task, yet most feed-f

MaskCoFT: Masked Co-Adaptive Fine-Tuning for Memory-Efficient MoE Inference

arXiv:2609.34077v1 Announce Type: new Abstract: Mixture-of-experts (MoE) language models often exceed the memory of a single GPU. Expert offloading keeps most experts in host memory and loads them on

Masking Frequent Tokens Sharpens Direct Preference Optimization

arXiv:2609.32445v1 Announce Type: cross Abstract: Direct Preference Optimization (DPO) aligns language models by optimizing over sequence-level sums of token-wise implicit reward differences. However,

MASTEST: A LLM-Based Multi-Agent System For Testing RESTful APIs

arXiv:2511.18038v2 Announce Type: replace-cross Abstract: Testing RESTful API is increasingly complicated but indispensable to quality assurance of cloud-native applications. This paper reports a mult

MDL-Calibrated Significance-Gain Pair Encoding: Replication-Aware Automatic Stopping for Subword Tokenization

arXiv:2609.31705v1 Announce Type: new Abstract: Byte-Pair Encoding (BPE) constructs subword vocabularies through greedy pair merging, but conventional BPE requires the number of merges or target vocab

Measure Learning at Steady State: A BIRD-SQL Formula 1 Case Study

arXiv:2609.31640v1 Announce Type: new Abstract: Continual Learning Bench scores learning as short-horizon gain versus a reset baseline and finds naive full-context ICL strongest among the memories it

MechBench: Can AI Scientific Agents Discover Mechanisms Beyond Phenomenal Laws?

arXiv:2609.35515v1 Announce Type: cross Abstract: Scientific discovery requires not only recovering mathematical laws that describe observable behavior, but also identifying the mechanisms that genera

Memory as a cache: Exact context reuse and deletion by construction

arXiv:2609.32395v1 Announce Type: cross Abstract: The KV cache of a transformer entangles every token's representation with its entire prefix: a passage encoded once cannot be reused under a different

Memory in the Sky: Low-Altitude Question Answering with Multi-Agent Memory Aggregation

arXiv:2609.35431v1 Announce Type: new Abstract: This paper studies low-altitude question answering (LAQA), in which distributed unmanned aerial vehicle (UAV) memories are aggregated at a ground server

MemTransfer: Benchmarking Memory Beyond Matched Experience in Embodied Decision-Making

arXiv:2609.32313v1 Announce Type: new Abstract: Memory lets an embodied agent reuse past experience, yet retaining useful information does not ensure that the agent can apply it when conditions change

MetaSampling: Making Frame Samplers Efficient for Long-Video Question Answering

arXiv:2609.33998v1 Announce Type: new Abstract: Frame selection is an important component of long-video question answering (VQA) with Multimodal Large Language Models (MLLMs). Existing frame-selection

Mind the Spike: Mechanisms and Brittleness of Visual Massive Activations in Large Vision-Language Models

arXiv:2609.32808v1 Announce Type: cross Abstract: Large vision-language models (LVLMs) inherit massive activations from their text-only bases: spikes where a few fixed hidden channels receive values t

Minimax-Optimality of Posterior Sampling for Reinforcement Learning

arXiv:2609.33246v1 Announce Type: new Abstract: Posterior sampling for reinforcement learning (PSRL) is one of the simplest and most effective exploration methods, but a basic question has remained op

MISHAP-Bench: A Hallucination Benchmark for Large Audio-Language Models

arXiv:2609.33893v1 Announce Type: new Abstract: Large audio-language models (LALMs) produce fluent responses about audio but often hallucinate by making plausible yet ungrounded claims. Existing audio

MixBench-TS: A Multivariate Time Series Forecasting Benchmark Where Channel Mixing Pays Off

arXiv:2609.32656v1 Announce Type: cross Abstract: Multivariate Time Series Forecasting (MTSF) models that mix information across channels assume that the past of one channel carries information about

MM-VeriRec: Failure-Guided Fusion for Verifiable Agentic Multimodal Recommendation

arXiv:2609.31718v1 Announce Type: cross Abstract: Images often carry recommendation constraints that text metadata only hints at. A movie may need to look dark, a product may need a minimal style, and

Model Casting and Low-Parameter Gating: Towards More Sparsely Activated FFNs

arXiv:2609.31975v1 Announce Type: new Abstract: This paper introduces model casting, a mid-training recipe that drastically sparsifies the activations within the Feed-Forward Network (FFN) layer. With

Models Designed to Forget: Machine Unlearning via Key Deletion

arXiv:2603.15033v4 Announce Type: replace Abstract: Machine unlearning for vision models is rapidly becoming a practical requirement, driven by privacy regulations, data errors, and the need to remove

MORPH: Self-Organising Multi-Robot Task Allocation via Neuroplasticity-Inspired Adaptive Topology

arXiv:2609.32745v1 Announce Type: new Abstract: Multi-robot task allocation (MRTA) in dynamic environments faces a fundamental tension: effective coordination requires learned structure, but that stru

MorphAtt: A Neuromorphic Accelerator for Efficient Multi-Head Attention Processing in Spiking Vision Transformers

arXiv:2609.33207v1 Announce Type: cross Abstract: Spiking Vision Transformers (SViTs) are developed as an energy-efficient alternative to conventional ViTs for computer vision tasks at the edge. Howev

Multi-Bitwidth Quantization for LLMs Using Additive Codebooks

arXiv:2606.12876v2 Announce Type: replace Abstract: As large language models (LLMs) are increasingly deployed across heterogeneous hardware with varying resource constraints, the ability to adaptively

Multi-Task Learning of Conditional Mean Operators: applications to dynamical systems and uncertainty quantification

arXiv:2609.35429v1 Announce Type: cross Abstract: Estimating conditional statistics and learning representations of a population of conditional distributions are central problems in many data-driven a

Multimodal LLMs Outperform Pathology Foundation Models in Cross-Domain Histological Similarity

arXiv:2609.32876v1 Announce Type: new Abstract: State-of-the-art pathology foundation models, trained on millions of histology tiles, can fail to preserve tissue similarity when comparisons cross slid

Muon Sublates the Edge of Stability in LLM Pretraining

arXiv:2609.34915v1 Announce Type: new Abstract: Muon is increasingly used for language-model pretraining, yet its large-step dynamics are not captured by the classical edge-of-stability (EoS) picture

Mycelium: A Generalizable Cross-Grid Multi-Task Model for Electrical Distribution Systems

arXiv:2609.33120v1 Announce Type: new Abstract: Electrical distribution grid operations require inference across heterogeneous networks from sparse, noisy, and incomplete time series measurements. In

NanoForecast v0.5: Competitive Time Series Forecasting Through Training Pipeline Optimization

arXiv:2609.31669v1 Announce Type: new Abstract: We present NanoForecast v0.5, a 6.5M-parameter forecaster that competes with models 31x its size (TimesFM, 200M parameters) after training pipeline fixe

Native Association: Confidence-Aware Human Perception in the Wild with a Foundation VLM

arXiv:2609.33450v1 Announce Type: new Abstract: Extracting who is where, on which team, wearing which number from a broadcast frame is typically done by stitching a detector, an OCR engine, and classi