Search: “benchmark”

TextVerifier: Seeing Beyond Recognition to Verify Chinese Text Rendering

arXiv:2610.19551v1 Announce Type: new Abstract: Text-to-image models increasingly generate readable Chinese text that still contains local rendering defects, such as a character with an extra or missi

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

The Agent Knew: Internal States Forecast Unauthorized Actions and Stop Them, and the Lies That Follow

DGX agent

arXiv:2610.17301v1 Announce Type: new Abstract: Language-model agents sometimes act beyond their authorization and then misreport what they did. On CAPE-Bench, matched agent tasks in which one fact de

model-releasesarxiv-cs-ai

15 Oct 2026

The Anatomy of a Blind Spot: What an LLM Judge Must Be Shown, and How Much

arXiv:2610.15938v1 Announce Type: new Abstract: Evaluation of text generated by large language models (LLMs) is increasingly delegated to LLMs acting as judges, yet what such a judge must be shown rem

researcharxiv-cs-ai

Titan-Llama: Retrofitting Pre-trained LLMs with External Neural Memory Modules

arXiv:2610.19210v1 Announce Type: new Abstract: Transformers are efficient to pre-train at scale because attention permits parallel computation across sequence positions, but long-context inference re

To Reason but Not to Stop: Understanding and Recovering Pruned Large Reasoning Models

arXiv:2610.19934v1 Announce Type: new Abstract: Weight pruning can degrade the accuracy of large reasoning models (LRMs), yet it remains unclear whether it damages reasoning itself. We find that much

researcharxiv-cs-cl

Token Utility for On-Policy Distillation via Local Trust-Region Optimization

arXiv:2610.18883v1 Announce Type: new Abstract: Token-selection methods for on-policy distillation (OPD) prioritize student-visited positions using diverse signals, including uncertainty, teacher-stud

safetyarxiv-cs-lg

TorchSpin: Differentiable Electron Paramagnetic Resonance Simulation in PyTorch

model-releasesarxiv-cs-lg

arXiv:2610.18924v1 Announce Type: cross Abstract: Interpreting electron paramagnetic resonance (EPR) spectra commonly requires repeated forward simulation and optimization of spin Hamiltonian paramete

Local Ai

Toward Fine-Grained Forgetting:Attribute Unlearning for Multimodal Large Language Models

arXiv:2608.01008v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) can memorize and disclose sensitive information, motivating machine unlearning to remove designated knowled

local-aiarxiv-cs-ai

Toward System One Clinical Intelligence: Parameter-Efficient Fine-Tuning and Counterfactual Explainability of Decision Foundation Models

arXiv:2610.15029v1 Announce Type: cross Abstract: Automated disease classification from patient-reported symptoms requires balancing the speed of lightweight discriminative models against the contextu

TRAP: Tool-result Replay Attack for Probing Misbinding in LLM Agents

arXiv:2610.16333v1 Announce Type: new Abstract: As AI agents increasingly act through external tools, tool results are no longer simple outputs but execution artifacts that must remain correctly assoc

TreeFlex: Agents Navigating Their Own Search Trees

arXiv:2610.20334v1 Announce Type: cross Abstract: A ReAct-style agent accumulates observations, intermediate results, and failed attempts in a single context. As this context grows, stale and off-path

Local Ai

TreeSpark: Learning to Draft the Trees You Verify for Speculative Decoding

arXiv:2610.20175v1 Announce Type: new Abstract: Parallel drafters such as DFlash and DSpark reduce proposal latency, but early mismatches limit how much of each block can be accepted. Retaining altern

local-aiarxiv-cs-cl

Hardware

Unlocking Structured Pruning for Reasoning Language Models via Attention-Preserving MLP Trimming

arXiv:2610.19121v1 Announce Type: cross Abstract: Structured pruning reduces LLM inference costs by removing model components to produce smaller dense models. However, existing methods can cause reaso

hardwarearxiv-cs-cl

Unlocking Token Rewards via Training-Free Reward Attribution

arXiv:2610.18704v1 Announce Type: new Abstract: In this paper, we propose an extremely efficient, training-free method to extract token-level reward signals directly from an existing deep reward model

v0.7.0

model-releasesllama-cpp-releases

Overview llama.cpp v0.7.0 adds the LiliCorr speculative decoding technique and MTP for GLM5-Next, a new GPU cache for MoE experts kept in host memory (--moe-cache-mib, up to ~2.2x decode speedup when

VASPO: Self-Distillation with Privileged Value Estimation

arXiv:2610.20516v1 Announce Type: new Abstract: On-policy self-distillation provides dense supervision for reasoning models via a privileged self-teacher. However, this supervision reflects local toke

safetyarxiv-cs-cl

VC-OPD: Visual Counterfactual On-Policy Distillation for Grounded Vision-Language Reasoning

arXiv:2610.14939v1 Announce Type: new Abstract: GRPO-style reinforcement learning has advanced vision-language reasoning, but its sequence-level rewards provide limited credit assignment for individua

safetyarxiv-cs-ai

Applications

Vela: User Behavior as Supervision Sparks LLM Intelligence

arXiv:2610.14709v1 Announce Type: cross Abstract: Large language models are increasingly deployed in applications that act on users' long-term behavior histories, from personal assistants to recommend

applicationsarxiv-cs-ai

VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?

arXiv:2605.06068v2 Announce Type: replace Abstract: For years, we have built LLM serving systems like any other critical infrastructure, with a single general-purpose stack hand-tuned over many engine

Video-Browser: Towards Open-web Agentic Video Browsing

arXiv:2512.23044v3 Announce Type: replace Abstract: Autonomous agents are shifting information seeking from passive retrieval toward proactive, open-ended research, yet this progress has largely bypas

Video Depth Anything: Consistent Depth Estimation for Super-Long Videos

arXiv:2501.12375v4 Announce Type: replace Abstract: Depth Anything has achieved remarkable success in monocular depth estimation with strong generalization ability. However, it suffers from temporal i

researcharxiv-cs-cv

Agents

VideoHarness-SRSI: Structured Recursive Self-Improvement for Video Understanding Harness

arXiv:2610.19424v1 Announce Type: new Abstract: LLM system capabilities depend not only on model parameters but also on the executable harnesses that organize information acquisition, tool use, reason

agentsarxiv-cs-cv

Virtual Head Attention

arXiv:2610.17959v1 Announce Type: cross Abstract: The design of attention mechanisms in long-context large language models faces an ``impossible triangle'' among KV cache size, training and prefill FL

Vision-Language Models Integrate Conflicting Evidence Like Biased Bayesian Observers

arXiv:2610.19407v1 Announce Type: new Abstract: Vision-language models must reconcile conflicting text and images. Prior work has characterized the resulting behavior in terms of modality biases--e.g.

W-JEPA: Wavelet Joint-Embedding Predictive Architecture for Wearable Activity Recognition

arXiv:2610.19411v1 Announce Type: new Abstract: A major challenge in wearable device activity recognition is the heterogeneity across datasets in sensor placement, channel count, sampling rate and sub

researcharxiv-cs-lg

We want to support the mathematics community in turning progress in AI into deeper human understanding and new discoveries. To support that …

We want to support the mathematics community in turning progress in AI into deeper human understanding and new discoveries. To support that goal, we are contributing 5 million to a new Simons Foundati

model-releasesopenai--x

What the Model Cannot See: Covert and Provable Text Invisibility in LLMs

arXiv:2610.15883v1 Announce Type: cross Abstract: Text wrapped in a trigger can be made behaviorally invisible to a language model: omitted from summaries and quotations, ignored in question answering

When Occupation-derived Skills Don't Help: On the Limits of Skill-Augmented Career Path Prediction

arXiv:2610.20186v1 Announce Type: new Abstract: Skills are widely assumed to play a central role in career transitions, motivating their use in career path prediction (CPP) systems. In this work, we c

When Should Enterprise Policy Decisions Be Learned, and When Should They Be Reasoned?

arXiv:2610.17643v1 Announce Type: new Abstract: An enterprise agent that answers under a written policy must choose an action before it says anything: answer, ask for a missing fact, ask for confirmat

When Skills Remain But Access Fails: Factorized Skill Memory for Continual Skill Internalization

arXiv:2610.15492v1 Announce Type: new Abstract: Skill internalization aims to convert externally specified procedural skills into persistent model capabilities, allowing language-model agents to execu

Which Metadata Matters for LLM Tool Selection in MCP? Evidence from Controlled Factor-Level Interventions

arXiv:2610.14334v1 Announce Type: new Abstract: In Model Context Protocol (MCP) systems, LLMs select tools from metadata rather than from their implementations. Names and descriptions summarize what a

Who's Behind the Text? Robust Model Authorship Attribution

arXiv:2610.19583v1 Announce Type: cross Abstract: As large language models (LLMs) are increasingly used to generate text, their misuse has become a growing societal problem, with implications for acad

Why Generate a Click? Plug-in Decision Heads for Web Agents

arXiv:2610.20107v1 Announce Type: new Abstract: Web agents built on vision language models decide every action by generating text such as click(12), although most of that text encodes a choice among a

researcharxiv-cs-cl

176 t/s decode and 6,450 t/s prefill with Q4 Qwen-3.8 Flash-Next on 2xR9700 + 64GB DDR5 using Deadcode's Radiance engine

model-releasesr-localllama

https://preview.redd.it/kudj7xii5ivh1.png?width=1508&format=png&auto=webp&s=07712e81310b81aceb2cf8f4d8a7c2eadeb0c304 Just built a new PC, full parts list is here. I am using Deadcode's Radiance infere

A CaLLM Judge is a Better Judge: Mitigating Bias in LLM-as-a-judge

arXiv:2610.18108v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as judges to predict which of two candidate answers a human evaluator would prefer, providing a scala

safetyarxiv-cs-lg

Local Ai

A Comprehensive Review on RNA Subcellular Localization Prediction

arXiv:2504.17162v2 Announce Type: replace Abstract: The subcellular localization of RNAs, including long non-coding RNAs (lncRNAs), messenger RNAs (mRNAs), microRNAs (miRNAs) and other smaller RNAs, p

local-aiarxiv-cs-cv

Agents

A Gated Catchment Hydrology Model (GCHM) for Physically-Interpretable Rainfall-Evapotranspiration-Runoff Simulation Across Diverse Hydroclimatic Regimes

arXiv:2610.14493v1 Announce Type: new Abstract: Recent advances in machine learning have improved rainfall-runoff prediction, yet most data-driven models remain difficult to interpret and do not expli

agentsarxiv-cs-lg

A Hybrid Kinematic GCN and Elastic Transformer for Efficient Indian Sign Language Recognition

arXiv:2610.14680v1 Announce Type: cross Abstract: Sign Language Recognition (SLR) presents complex challenges due to variable signing speeds and spatial intricacies. Specifically, Indian Sign Language

researcharxiv-cs-cv

A Multi-Task Deep Learning Framework for Pose-Invariant Face Recognition and Facial Landmark Localization

arXiv:2610.16849v1 Announce Type: new Abstract: Face recognition systems often experience performance degradation under large pose variations due to changes in facial appearance and visibility of disc

A Reinforcement Learning Framework for Safety-Aware Haptic Shared Control in Surgical Robotics

model-releasesarxiv-cs-ro

arXiv:2610.15422v1 Announce Type: new Abstract: Haptic shared control in surgical robotics combines human-side and robot-side forces at a shared end effector while preserving operator involvement. How

A text-guided vision model for enhanced recognition of small instances

arXiv:2602.19503v2 Announce Type: replace Abstract: As drone-based object detection technology continues to evolve, the demand is shifting from merely detecting objects to enabling users to accurately

ACED: Reference-Anchored Aggregation of Natural-Language and Executable Evidence for Mathematical Reasoning

arXiv:2610.16635v1 Announce Type: new Abstract: Natural-language self-consistency and executable programs provide complementary test-time evidence for mathematical reasoning, yet either channel may su

researcharxiv-cs-lg

Across Viewpoints and Time: Trajectory-Generalizable Physical Attacks on World Models and Video Understanding Models

arXiv:2610.15343v1 Announce Type: new Abstract: World models and video understanding models are increasingly deployed in dynamic visual environments, yet their robustness to physical adversarial threa

safetyarxiv-cs-cv

AdaGamma: State-Dependent Discounting for Temporal Adaptation in Reinforcement Learning

arXiv:2605.06149v2 Announce Type: replace Abstract: The discount factor in reinforcement learning controls both the effective planning horizon and the strength of bootstrapping, yet most deep RL metho

safetyarxiv-cs-lg

AeroChem: Closed-loop Physics-Informed State Space Modeling for Long-term Chemically-Reactive Air Quality Forecasting--Extended Version

model-releasesarxiv-cs-lg

arXiv:2610.16630v1 Announce Type: new Abstract: Accurate long-term air quality forecasting is challenging due to long-range transport, meteorological variability, and nonlinear chemical reactions unde

AERR-Nav: Adaptive Exploration-Recovery-Reminiscing Strategy for Zero-Shot Object Navigation

arXiv:2603.17712v1 Announce Type: cross Abstract: Zero-Shot Object Navigation (ZSON) in unknown multi-floor environments presents a significant challenge. Recent methods, mostly based on semantic valu

researcharxiv-cs-lg

Agaram 2.5B: The First Tamil-First Language Model Trained from Scratch with Tool Calling. A Playbook for Under-Represented Languages

model-releasesarxiv-cs-lg

arXiv:2610.18420v1 Announce Type: cross Abstract: We trained Agaram 2.5B, a 2.5-billion-parameter Tamil-English language model, from scratch on about 50 billion tokens (58% Tamil, the rest English, ma

Tutorials

Akule: Fast and Scalable Multi-Robot Motion Planning via Sparse Interaction Diffusion

arXiv:2610.14248v1 Announce Type: new Abstract: Diffusion models provide expressive trajectory priors for multi-robot motion planning (MRMP), but scaling them to large teams requires coordinating indi

tutorialsarxiv-cs-ro