Search: “models”

Measuring Data Diversity with Embeddings: A Taxonomy and Benchmark for Diversity Measures

arXiv:2610.24888v1 Announce Type: cross Abstract: Diversity is a critical property of text corpora, with direct implications for training data curation, test set design, and model generalization. Howe

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

miniTE: Diagnosing and Bridging the Execution Gap in Native NVFP4 Reinforcement Learning

DGX agent

arXiv:2610.23928v1 Announce Type: new Abstract: Native NVFP4 promises to accelerate reinforcement learning (RL) for large language models (LLMs), but achieving stable learning across separate training

model-releasesarxiv-cs-lg

20 Oct 2026

Nash-MOPD: Optimizing Task Trade-offs in Multi-Teacher On-Policy Distillation

arXiv:2610.26124v1 Announce Type: new Abstract: Multi-teacher on-policy distillation (MOPD) aggregates the capabilities of multiple domain-expert teachers into a single student model. However, all dom

Negation without Confusion: Decomposing Bias and Recomposing Style in Text Embeddings for Negation-aware Image Generation

arXiv:2610.25645v1 Announce Type: new Abstract: Negation-aware image generation remains challenging for text-to-image diffusion models, which often struggle with negation due to implicit linguistic bi

safetyarxiv-cs-cv

OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory

arXiv:2610.26860v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) have made rapid progress on visual reasoning, yet many tasks require evidence that is difficult to obtain thr

agentsarxiv-cs-cv

OV-PEAR: Observation-Verified Pose-Scale Estimation and Reconstruction for Unknown Industrial Manipulation Objects

arXiv:2610.25840v1 Announce Type: new Abstract: Fine industrial robotic manipulation requires 6D pose, scale, and an operation-ready 3D model. Existing industrial pose estimation relies on CAD models,

safetyarxiv-cs-ro

PACE: Pareto-Adaptive Compression for Efficient Native MLLMs

arXiv:2610.24005v1 Announce Type: new Abstract: Native multimodal large language models (native MLLMs) jointly encode visual and linguistic sequences within a unified autoregressive Transformer, emerg

Paloma: Phase-Conditioned Residual Modulation for Time Series Forecasting

arXiv:2610.25392v1 Announce Type: new Abstract: Explicitly modeling periodicity has proven to be an efficient approach for time series forecasting, where the residual, obtained after removing the peri

R2-GP: Variance-Calibrated Priors for Additive Gaussian Processes

arXiv:2610.26437v1 Announce Type: cross Abstract: We develop R2-GP, an R2D2-type prior for Gaussian process (GP) generalised additive models (GAMs) that calibrates explained variance (R^2) for Matern

Regime-Aware Interpretable Ranking for Sector Rotation

arXiv:2610.26456v1 Announce Type: cross Abstract: Sector rotation is a common decision in benchmark-relative portfolio management, where portfolio managers rank sectors for overweight and underweight

RestoRect: Degraded Image Restoration via Latent Rectified Flow & Feature Distillation

arXiv:2509.23480v3 Announce Type: replace Abstract: Current approaches for restoration of degraded images face a trade-off: high-performance models are slow for practical use, while fast models produc

safetyarxiv-cs-cv

Severity-Channel Conditioning: An Injection-Level Study for Consumer-Device Deblurring

arXiv:2610.24599v1 Announce Type: cross Abstract: Motion-deblurring models lose substantial accuracy when transferred from professional cameras to smartphone-derived imagery. This paper studies whethe

SituHOI: Scalable Capture of Scene-Situated Full-Body Human-Object Interaction

arXiv:2610.23980v1 Announce Type: new Abstract: Modeling human-object interaction (HOI) in real environments requires jointly representing whole-body motion, manipulated objects, and scene geometry, y

SPRINT: Stochastic Performative Prediction With Variance Reduction

arXiv:2509.17304v3 Announce Type: replace Abstract: Performative prediction (PP) is an algorithmic framework for optimizing machine learning (ML) models where the model's deployment affects the distri

Sufficient Frame Subsets for Video Understanding

arXiv:2610.25084v1 Announce Type: new Abstract: Understanding when during a video a model makes its critical observations is essential for interpreting video-understanding systems. However, existing a

Super-resolution as Spatial Downscaling: Coherence, Structure and Task Utility in Earth Observation

arXiv:2610.23972v1 Announce Type: new Abstract: Learned satellite super-resolution is commonly assessed by pixel fidelity and potential downlink savings, but these criteria do not establish whether th

The Missing Compliance Class: LLM Graders That Collapse Jailbreak Success, and a Free Check

arXiv:2610.24872v1 Announce Type: cross Abstract: Jailbreak papers report an attack success rate, the share of responses that deliver the harmful request, and read it as a property of the target model

safetyarxiv-cs-lg

Token-Set Choice Confounds POPE: A Systematic Audit of Yes/No Extraction in VLM Hallucination Evaluation

arXiv:2610.25964v1 Announce Type: new Abstract: We set out to reproduce a recent hallucination-mitigation result on the POPE benchmark. The LLaVA-1.5-7B baseline we measured, 82.21 F1 on the adversari

TTCS: Test-Time Curriculum Synthesis for Self-Evolving

arXiv:2601.22628v2 Announce Type: replace Abstract: Test-Time Training offers a promising way to improve the reasoning ability of large language models (LLMs) by adapting the model using only the test

UltraData-Code: From Raw Repositories to Algorithmically Dense and Task-Oriented Code Data

arXiv:2610.23960v1 Announce Type: new Abstract: Code generation has become a core capability of large language models (LLMs), and code data is central to the pre-training process that develops it. As

VERDICT: Visual Extraction-Result Decoder with Integrated Confidence and Traceable Evidence

arXiv:2610.23933v1 Announce Type: new Abstract: Enterprises run schema-guided structured field extraction on documents with frontier multimodal models, and accuracy keeps improving as new models and s

Visual Scaffolding: Toward Native Multimodal Reasoning

arXiv:2610.25247v1 Announce Type: new Abstract: The success of chain-of-thought (CoT) in complex reasoning can be attributed to a textual scaffolding effect: explicit intermediate results can be exami

VLA-Rolling: Fewer Denoising Steps, Better Control via Adaptive Action Reuse

model-releasesarxiv-cs-ro

arXiv:2610.25043v1 Announce Type: cross Abstract: Flow-matching vision-language-action (VLA) models generate an action chunk by iteratively denoising Gaussian noise, but typically execute only a short

Weighted State Souping for Efficient Long-Context Retrieval

arXiv:2610.25621v1 Announce Type: new Abstract: State space models (SSMs) combine fast inference and fixed state size with competitive language-modeling performance. Recently, state souping has been p

agentsarxiv-cs-lg

What Matters for NVFP4 Training? A Comparative Study of Low-Precision Pretraining Recipes

arXiv:2610.24343v1 Announce Type: cross Abstract: Training large language models directly in 4-bit floating-point (FP) formats promises substantial improvements in throughput and energy efficiency. Wh

Where and Why Search-Augmented LLMs Fail When Reasoning over Rapidly Evolving Knowledge

arXiv:2610.25323v1 Announce Type: cross Abstract: As the world changes, new information is published while older records remain online. This complicates reasoning for large language models (LLMs), whi

Whistle and Wrong Lyrics: Synthetic Queries Are Not Enough for Real-World Query-by-Melody

arXiv:2610.25371v1 Announce Type: cross Abstract: People look up a song by humming it, but also by whistling it or singing it with the wrong words. Published query-by-humming systems and benchmarks fo

A lot of professional work around documents can be arbitrarily challenging to automate. It not only requires deep visual perception/OCR arou…

model-releasesjerry-liu--x

A lot of professional work around documents can be arbitrarily challenging to automate. It not only requires deep visual perception/OCR around complex entities (diagrams/charts), but also requires a r

A Minimal ReLU Classifier on Non-Linearly Separable Data: Convergence and Implicit Bias at and beyond the Edge of Stability

arXiv:2610.23007v1 Announce Type: new Abstract: Gradient descent (GD) with large learning rates can successfully optimize neural networks even though classical smoothness-based learning rate condition

Anytime Contracts for AI Services

arXiv:2610.22781v1 Announce Type: cross Abstract: AI services are prone to moral hazard, as model providers can silently route requests to cheaper, lower-quality models, and the inherent stochasticity

agentsarxiv-cs-lg

Beyond Marginal Influence: Interaction-Aware Memory Selection for LLM Agents

arXiv:2610.23266v1 Announce Type: cross Abstract: LLM-agent memory selection often uses retrieval rankings based on record relevance or individual influence. Such rankings cannot identify sufficient s

Beyond Surface Cues: Robust LLM-Generated Korean Text Detection

model-releasesarxiv-cs-cl

arXiv:2610.22940v1 Announce Type: new Abstract: As large language models (LLMs) generate increasingly fluent text, the need for reliable detection of LLM-generated content has grown substantially. How

Broad and Efficient PII Detection with Unified Labels

model-releasesarxiv-cs-cl

arXiv:2610.22597v1 Announce Type: new Abstract: Personally Identifiable Information (PII) detection models must efficiently find identifiers across diverse contexts while maintaining a low false-posit

Can the Parts Fool the Test? Counterfactual Pairing Cycles for Relational OOD

arXiv:2610.23248v1 Announce Type: new Abstract: Relational out-of-distribution errors are easy to describe and hard to evaluate cleanly. An image may be natural, a caption may be fluent, and all menti

COMET: Decoupled Distillation, Routing, and Capacity Control for Task-Agnostic Continual Vision--Language Learning

arXiv:2610.22907v1 Announce Type: cross Abstract: Vision-language models (VLMs) such as CLIP retain zero-shot recognition ability, but their open-world use often requires task-agnostic continual learn

Deep LLM layers expose passage-level context in EEG during silent reading

arXiv:2610.23638v1 Announce Type: new Abstract: In a previous report we showed that a decoder can recover open-vocabulary word-level information from 19-channel EEG recorded during silent reading, by

Diversity Without Labels: Task-Agnostic Coreset Selection Across Perception and Action

arXiv:2610.23642v1 Announce Type: new Abstract: While data collection for model training is relatively easy, curating high-quality datasets from large collections is inherently challenging. Coreset es

Do Multi-Agent LLMs Explore Each Other?

arXiv:2607.11250v2 Announce Type: replace-cross Abstract: Exploration is essential for reliable autonomy in multi-agent systems, yet it remains unclear whether large language model (LLM) agents can ex

agentsarxiv-cs-ai

Eliciting Specifications: When Does Interactive Clarification Help?

arXiv:2610.22382v1 Announce Type: cross Abstract: LLMs can generate Lean specifications from task descriptions, but those descriptions are often underspecified or ambiguous and may omit verifier-relev

Exploring The Empirical Privacy Landscape of Synthetic Data

arXiv:2610.22659v1 Announce Type: cross Abstract: Methods for training language models on sensitive data differ in how they balance capability gains against the risk of leaking private information. In

researcharxiv-cs-cl

GatewayBench: Unified Capability and Security Evaluation Platform for MCP Gateways

arXiv:2610.23449v1 Announce Type: new Abstract: AI agents have increasingly interacted with external environments by invoking tools, with Model Context Protocol (MCP) emerging as a standard interface

HardEval: An Execution-Verified Benchmark for AI-Driven Security Hardening and Compliance Remediation

arXiv:2610.23685v1 Announce Type: cross Abstract: Implementing Security Technical Implementation Guide (STIG) controls remains a demanding operating-system administration task. Each natural-language s

How Finite-Rank Bottlenecks Shape the Low-Rank Adaptation Landscape

arXiv:2610.23343v1 Announce Type: new Abstract: Low-rank adaptation (LoRA) has become the standard method for parameter-efficient fine-tuning of large pretrained models, yet theoretical explanations f

LexPilot: Benchmarking the Proactive Interaction Ability of Legal LLMs with Laypeople in Chinese Civil Case Consultation

arXiv:2610.23141v1 Announce Type: new Abstract: Large Language Models (LLMs) have greatly assisted legal professionals. However, our empirical studies reveal that, when faced with underspecified consu

Look Back, Correct Forward: Error Feedback Calibration for Time Series Forecasting

arXiv:2610.22872v1 Announce Type: new Abstract: Time series forecasting models are increasingly deployed in continuously evolving environments. After a forecast is issued, future observations graduall

Many Biases, One Circuit: Post-Training Tunes a Pretrained Cue-Routing Mechanism

arXiv:2610.22937v1 Announce Type: cross Abstract: Language models often abandon a correct answer when the prompt contains a cue, a hint toward a wrong option such as 'I think the answer is B.' Sycopha

safetyarxiv-cs-ai

MS-Eval: Benchmarking Sub-Agent Delegation and Orchestration in MCP-Intensive Working Scenarios

arXiv:2610.22869v1 Announce Type: new Abstract: LLM agents are increasingly integrated into real-world works, yet they must handle diverse domain-specific tools and complex multi-sub-task problems tha

MTEdit: Packed Trajectories Beat Rollouts for Multi-Turn Image Editing

arXiv:2610.23751v1 Announce Type: new Abstract: Multi-turn image editing aims to modify an image according to a sequence of instructions, making it better aligned with human-AI collaborative creation