Search: “evaluation”

arXiv:2609.38830v1 Announce Type: new Abstract: Sparse attention is widely used to accelerate long-context inference in modern large language models (LLMs), but its input-dependent execution behavior

hardwarearxiv-cs-lg

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Staying on Task: Testing the Foundations of Long-Horizon Agent Reliability

DGX agent

arXiv:2609.38712v1 Announce Type: new Abstract: Long-horizon agentic workflows require models to sustain repeated state-dependent actions all while the context grows, sub-task complexity changes, and

safetyarxiv-cs-ai

1 Oct 2026

Strong Multilingual Privacy Tagging at Encoder Speed

arXiv:2609.38630v1 Announce Type: new Abstract: Privacy redaction must remove personal information while preserving relationships expressed in text. We develop a multilingual named-entity tagger with

local-aiarxiv-cs-cl

Synthesis Without Training: An Inference-Only Pipeline for Tabular, Temporal, and Relational Synthetic Data

arXiv:2609.38414v1 Announce Type: new Abstract: Synthetic data generation is dominated by the fit-then-sample paradigm: a generative model is trained on a private dataset and then sampled from. Despit

agentsarxiv-cs-lg

Tacit-TTS: From Autoregressive Decoding to Masked Prediction for Efficient Transcript-Free Voice Cloning

arXiv:2609.38658v1 Announce Type: cross Abstract: TTS systems with autoregressive semantic modeling have demonstrated strong zero-shot voice cloning performance and rich expressive variation, but thei

researcharxiv-cs-cl

TALK-Dem: Benchmarking Embodied Task Planning under Dementia-Associated Communication Patterns

model-releasesarxiv-cs-cl

arXiv:2609.38371v1 Announce Type: cross Abstract: Existing LLM-driven robot task planners rely on a taken-for-granted assumption of an ideal user whose instructions are clear, complete, and task-focus

Targeted Retrieval, Compact Representations: How CoT Reasoning Improves Long-Context Counting

arXiv:2609.38958v1 Announce Type: new Abstract: Large language models (LLMs) have been rapidly improving in long-context tasks, powered by Chain-of-Thought (CoT) reasoning. However, the internal mecha

TRACE: Target-Aware Retrieval, Attributed Evidence, and Contract-Constrained Extraction for LitTraceQA

arXiv:2609.38861v1 Announce Type: new Abstract: Finding a relevant paper is not the same as producing a verifiable answer from it. LitTraceQA requires canonical paper identifiers, exact evidence at th

researcharxiv-cs-cl

TRACE: Trajectory Selection for Parallel Scaling of Search Agents

arXiv:2609.39912v1 Announce Type: cross Abstract: Parallel search may generate a correct answer that final-answer voting fails to select. We formulate this consolidation stage as trajectory selection

agentsarxiv-cs-ai

Trust the Critic More

arXiv:2609.39247v1 Announce Type: cross Abstract: Standard language model RL algorithms credit every token of a long rollout with the same advantage determined by the terminal reward. Actor-critic met

local-aiarxiv-cs-ai

Trustworthy Runtime Error Healing in Real-World Repositories: A Benchmark and Guardrail

arXiv:2609.39086v1 Announce Type: cross Abstract: Runtime error healing lets a crashed program continue by generating code that repairs its live runtime state. Recent work shows that LLMs can generate

TSMD: Temporal-Stream Modality Dropout for Robust Video Highlight Detection

arXiv:2609.39051v1 Announce Type: new Abstract: Existing multimodal video highlight detectors typically assume that visual, audio, and textual streams are continuously available. In practice, however,

Two-Step Data Augmentation for Masked Face Detection and Recognition: Turning Fake Masks to Real

arXiv:2512.15774v5 Announce Type: replace Abstract: The absence of large-scale masked face datasets challenges masked face detection and recognition. We propose a two-step generative data augmentation

Uncertainty-Normalized Margins for Direct Preference Optimization

model-releasesarxiv-cs-lg

arXiv:2609.38647v1 Announce Type: new Abstract: Direct preference optimization (DPO) models binary preferences through a Bradley-Terry model with a common noise scale, without explicitly accounting fo

Safety

Unveiling the Value of Motion for Cinematic Camera Trajectories

arXiv:2609.38683v1 Announce Type: cross Abstract: Cinematic camera motion is a fundamental storytelling tool, defined not only by where the camera is positioned in the scene, but also by how it moves

safetyarxiv-cs-ai

URBAN-SPIN: A street-level bikeability index to inform design implementations in historical city centres

arXiv:2602.10124v2 Announce Type: replace-cross Abstract: Cycling is reported by an average of 35% of adults at least once per week across 28 countries, and as vulnerable road users directly exposed t

Uruqi: Learning Spatial Cognition from Visual Experience

model-releasesarxiv-cs-cv

arXiv:2609.39195v1 Announce Type: new Abstract: Spatial intelligence requires maintaining a coherent understanding of the world as the embodied agent moves. Like humans, the agent must use its own mot

Values as Style: Disentangling Values from Semantics with One-Way Mixing for Low-Damage LLM Steering

arXiv:2609.39701v1 Announce Type: new Abstract: Value steering should change an LLM's normative priorities while preserving the scenario, facts, and task constraints underlying its answer. Conventiona

ViLegalExpert: A Large-Scale Benchmark for Vietnamese Legal Retrieval and Question Answering from Real-World Consultations

model-releasesarxiv-cs-cl

arXiv:2609.39189v1 Announce Type: new Abstract: Trustworthy Legal AI requires systems that can answer legal questions while grounding their responses in authoritative sources. However, existing Vietna

Visualizing Distribution Coverage in Generative Diffusion Models

arXiv:2609.38853v1 Announce Type: new Abstract: Diffusion distillation is widely adopted to accelerate sampling, and the resulting few-step models are broadly believed to match or even surpass their m

What Can Component-Replacement Evidence Establish? A Critical Scoping Review of Local Decisions in LLM Agents

arXiv:2609.39989v1 Announce Type: new Abstract: Background. A component replacement in a language-model agent changes an execution trajectory, potentially altering later observations, resource use, an

local-aiarxiv-cs-ai

What Limits Recursive Reasoning Models: Optimization, Architecture and Test-Time Scaling

arXiv:2609.39967v1 Announce Type: cross Abstract: Recursive reasoning models apply a small shared Transformer block many times to refine a latent state. This gives them large effective depth with few

When Can Text Replace Vision? Structural Bottlenecks in Diagram Reasoning

arXiv:2609.39142v1 Announce Type: new Abstract: Can structured text replace vision for diagram reasoning? A wrong answer after textualization can arise because the representation omits information the

Tutorials

When Correct Memory Goes Wrong: Fuzzing Persistent Memory Use in LLM Agents

arXiv:2609.38275v1 Announce Type: cross Abstract: Persistent memory helps LLM agents carry information across long interactions, but correct memory can still be used incorrectly when queries change or

tutorialsarxiv-cs-ai

When Reasoning Goes Astray: Attention Dynamics of Uncontrolled Reasoning

arXiv:2609.38817v1 Announce Type: new Abstract: Large reasoning models (LRMs) improve performance on complex tasks through extended reasoning, yet the same process can degenerate into redundant verifi

When Scientific Contradictions Are Lost in Translation

arXiv:2609.38621v1 Announce Type: new Abstract: Two scientific findings can disagree without contradicting each other. Determining whether they conflict requires knowing whether they describe comparab

Where MLLMs Fail and Why: Causal Task Decomposition for Capability Failure Diagnosis

arXiv:2609.38851v1 Announce Type: cross Abstract: End-to-end accuracy on compositional tasks records how often MLLMs fail, but cannot distinguish whether a failure reflects an intrinsic deficit in the

Where Scientific Search Agents Fail: Decision-Checkpoint Auditing of Exposure and Inspection Attempts

arXiv:2609.38670v1 Announce Type: new Abstract: Final-answer accuracy does not reveal whether a scientific-search agent failed to encounter a target paper, attempt to inspect it, or return an accepted

Safety

Whose Voice Survives the Summary? A Voice-Retention Audit of LLM Employee Listening

arXiv:2609.38818v1 Announce Type: new Abstract: Organizations increasingly route employee feedback to leaders through large language model (LLM) summaries, an unaudited layer that silences already-spo

safetyarxiv-cs-ai

WorldAuditBench: Interactive 3D World Auditing with Multimodal Agents

arXiv:2609.40325v1 Announce Type: new Abstract: As interactive 3D worlds are increasingly used to study intelligent behavior, it becomes important to develop efficient pipelines for identifying anomal

You're Hired: Strategic Model Selection for LLM Collaboration

arXiv:2609.38816v1 Announce Type: new Abstract: While multi-agent and model collaboration algorithms gain traction to combine the strengths of diverse Large Language Models (LLMs), existing systems re

agentsarxiv-cs-cl

Hardware

AI as a Compiler: Compiling Triton kernels without the Triton compiler

arXiv:2609.36800v1 Announce Type: new Abstract: Compiler backends are expensive to build and maintain as programming models, workloads, and accelerators evolve. We investigate whether large language m

hardwarearxiv-cs-ai

Alpha Diffusion Language Models: Factorization Alone Is Not the Problem

arXiv:2609.38066v1 Announce Type: new Abstract: Discrete diffusion language models can generate multiple tokens in parallel, but reducing the number of denoising steps can lead to inconsistent predict

An Efficient Machine Learning Approach for Degradation Forecasting in AEM Water Electrolysis

arXiv:2609.37941v1 Announce Type: new Abstract: This study provides a data-driven analysis of a novel dataset of single-cell Anion Exchange Membrane water electrolyzers (AEMWE), operated under constan

Applications

APM-Bench: Benchmarking Cross-session Persistent Memory for Egocentric Streaming Video Assistants

arXiv:2609.37559v1 Announce Type: new Abstract: To serve as real-world personal assistants, streaming video models need persistent memory that retains past experiences for later use. Yet existing stre

applicationsarxiv-cs-cv

ARGOS: Reinforcement Learning-Driven Multidimensional Elasticity for Service Orchestration in the Computing Continuum

arXiv:2609.37085v1 Announce Type: cross Abstract: Data-intensive services in the Computing Continuum must balance analytics quality, resource usage, and cost across heterogeneous nodes with limited an

ASCEND: Personal AI Agents for Autonomous Scientific Computing Across HPC Clusters and GPU Workstations

arXiv:2609.32868v2 Announce Type: replace-cross Abstract: Traditional scientific computing requires researchers to translate computational intent into environment configuration, resource requests, and

au-Multilingual: Benchmarking Voice Agents Across Languages

arXiv:2609.35820v1 Announce Type: cross Abstract: English-only benchmarks expose only a narrow slice of voice-agent behavior. We introduce au-Multilingual, extending au-Voice to Spanish, Brazilian Por

agentsarxiv-cs-ai

AutoMark: Enabling Autoresearch to Discover Better LLM Watermarks

model-releasesarxiv-cs-lg

arXiv:2609.37310v1 Announce Type: cross Abstract: With LLM watermarking being deployed commercially and now required by regulations, improving its reliability and effectiveness has become crucial. Yet

BarcodeMAE+: Rethinking Masked Pretraining and Global Representations for DNA Barcode Foundation Models

arXiv:2609.35877v1 Announce Type: cross Abstract: Many DNA foundation models are pretrained by masking parts of a sequence and asking the model to reconstruct them. Standard masked pretraining exposes

Better Nearest Neighbor Graph Indices via (Efficient) LLM-Guided Pruning

arXiv:2609.36359v1 Announce Type: new Abstract: Graph-based approximate nearest neighbor search (ANNS) is widely used for large-scale semantic search. Its indices are constructed primarily based on ge

Beyond a single latent space: a dual-latent world model for long-horizon planning

arXiv:2609.37644v1 Announce Type: new Abstract: Latent world models often struggle with long-horizon planning despite accurate short-term predictions. Recursive rollouts accumulate errors, while dista

local-aiarxiv-cs-ai

Beyond Legibility: Benchmarking Visual Text Rendering and In-Place Editing in Unified Video Generation

model-releasesarxiv-cs-cv

arXiv:2609.36598v1 Announce Type: new Abstract: A video can exhibit convincing motion and photorealism yet fail immediately when visual text collapses. Unlike generic scene content, visual text is unf

Beyond Lip Sync: Reference-Grounded Oral Refinement for Audio-Driven Portrait Animation

model-releasesarxiv-cs-cv

arXiv:2609.38019v1 Announce Type: new Abstract: We present RGOR (Reference-Grounded Oral Refinement), an audio-driven lip-sync framework that renders the mouth of the specific person being dubbed rath

Hardware

BiFE: Search-Efficient Discovery of CPU-Only Branching Policies via LLM-based Bi-Fidelity Evolution

arXiv:2609.36735v1 Announce Type: new Abstract: In branch-and-bound (B&B) for mixed-integer linear programming (MILP), branching variable selection critically impacts efficiency. Existing neural branc

hardwarearxiv-cs-ai

CADENCE: A Confidence-Adaptive Dual-Expert Network for Fast and Accurate Time Series Classification

arXiv:2609.35929v1 Announce Type: new Abstract: Time series classification (TSC) exhibits a sharp trade-off between accuracy and computational scalability. Meta-ensembles like HIVE-COTE 2.0 reach stat

Can Agents Design Libraries for Agents?

arXiv:2609.36730v1 Announce Type: new Abstract: Agents increasingly build on code written by other agents, and they reimplement rather than reuse, growing the codebases later agents must work in. To m

Challenges and Solutions for Bandits in the Wild: Warm-Started Mixture Bandits for Cross-Cohort Slate Recommendation

arXiv:2609.37800v1 Announce Type: cross Abstract: Many recommender services repeatedly encounter cold-start cohorts, where new users arrive with little or no interaction history. This creates two chal