Search: “inference”

When Do LLMs Admit Their Mistakes? Understanding The Role Of Model Belief In Retraction

arXiv:2505.16170v4 Announce Type: replace Abstract: We study the internal mechanisms that govern when LLMs choose to retract wrong answers, i.e., spontaneously and immediately acknowledge errors in th

tutorialsarxiv-cs-cl

10 Aug 2026

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Tutorials

Why Knowing Both Hops Is Not Enough: Understanding Two-Hop Generalization in Language Models

DGX agent

arXiv:2608.07261v1 Announce Type: new Abstract: Large language models (LLMs) can solve complex multi-hop problems yet exhibit puzzling failures on simple two-hop queries: although a model may correctl

tutorialsarxiv-cs-cl

10 Aug 2026

ZIPBrain: Can EEG Foundation Models Be Faster, Locally Deployable, but Accurate?

arXiv:2608.07033v1 Announce Type: new Abstract: This work investigates whether Electroencephalograph (EEG) foundation models (EFMs) can be made faster and locally deployable without sacrificing accura

10 Aug 2026

Beyond Sequence Order: Syntax-Informed Positional Embeddings for Transformers

arXiv:2608.06111v1 Announce Type: cross Abstract: Positional embeddings (PE) in Transformers encode token distance and order but are largely agnostic to extit{syntactic structure}. We introduce extbf{

BioKD: Selective Physiology-to-Video Knowledge Distillation via Reliability Gate for Emotion Recognition

arXiv:2608.06023v1 Announce Type: new Abstract: To address the limitations of video-based emotion recognition under ambiguous or socially masked behavioral cues, as well as the poor deployability of p

tutorialsarxiv-cs-lg

Clinician input steers AI toward accurate and harmful recommendations

model-releasesarxiv-cs-lg

arXiv:2603.14158v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are entering clinical workflows, yet evaluations rarely assess how clinician reasoning shapes model behavior duri

Continual Learning in Transition

arXiv:2608.06216v1 Announce Type: cross Abstract: Classical continual learning (CL) has primarily focused on enabling models to update and retain knowledge through parameter-centric mechanisms, e.g.,

Applications

DTRNet: Dual Text-Radical Decoding for Handwritten Chinese Text Recognition with Faked Character Detection

arXiv:2608.05848v1 Announce Type: new Abstract: In K-12 educational scenarios, handwritten Chinese text recognition should not only transcribe student writing, but also detect faked characters. Howeve

applicationsarxiv-cs-cv

From Passive Mirrors to Active Agents: Holonic Digital Twins for Physical AI over Networks

arXiv:2608.06227v1 Announce Type: cross Abstract: Despite advances in artificial intelligence (AI) across multiple sectors, today's AI tools, including deep learning and generative AI, still fail when

In-Context Forcing: Uncovering Context Effects in Autoregressive Video Diffusion

arXiv:2608.05237v1 Announce Type: cross Abstract: Current few-step autoregressive video diffusion models depend on previous fully denoised clean frames as context for all denoising steps of the curren

Applications

LiteKD-Net: Lightweight Knowledge-Distilled Network for Mobile Image Denoising

arXiv:2608.05739v1 Announce Type: new Abstract: Mobile image denoising requires both good restoration quality and low computational cost. In addition, it's annoying to collect large-scale LQ-GT clean

applicationsarxiv-cs-cv

Coupled Continuous-Discrete Generation for Scene Text Image Super-Resolution

model-releasesarxiv-cs-cv

arXiv:2608.04525v1 Announce Type: new Abstract: Scene text image super-resolution (STISR) aims to recover visually plausible appearance while preserving character semantics from degraded inputs. Exist

DreamWAM: Beyond RGB Future Prediction for World Action Models

model-releasesarxiv-cs-ro

arXiv:2608.04996v1 Announce Type: new Abstract: World Action Models (WAMs) learn action-relevant representations by predicting how the observed world will evolve. Most existing WAMs define this future

Enhancing Trustworthy Clinical Diagnosis Decision-Making in Large Language Models via Etiology-Aware Attention Supervision

arXiv:2508.00285v2 Announce Type: replace Abstract: Objective: Large Language Models (LLMs) have demonstrated strong capabilities in medical text understanding and generation. However, their trustwort

FM4WiFi: Flow Matching for Multi-AP Coordination in Dense Deployments of Beyond Wi-Fi 8 Networks

arXiv:2608.04050v1 Announce Type: cross Abstract: Wi-Fi networks are moving beyond random channel access toward tightly coordinated operation across access points (APs), a shift reflected in Wi-Fi 8's

researcharxiv-cs-lg

Hardware

GASP: GPU-Accelerated Safe Planner for Real-Time Collision-Aware Motion Generation with Latent Trajectory Sampling

arXiv:2608.04612v1 Announce Type: new Abstract: We present GASP, a GPU-Accelerated Safe Planner for real-time, collision-aware joint-space motion generation in known environments. GASP combines a clam

hardwarearxiv-cs-ro

HiSC: Hierarchical Spatial Clustering Token Compression for Efficient 3D Scene Understanding

arXiv:2608.04610v1 Announce Type: new Abstract: 3D vision-language models (3D VLMs) enable spatial reasoning over multi-view scenes but suffer from substantial token redundancy due to duplicated obser

researcharxiv-cs-cv

Manipulation-Proof Oblivious Audits against Deceptive Model Providers

arXiv:2608.04365v1 Announce Type: new Abstract: Audits have emerged as a critical instrument for algorithmic governance, providing a mechanism for external scrutiny and governance of machine learning

safetyarxiv-cs-lg

Reading Between the Frames: Interpreting Implicit and Non-literal Meaning in Social Media Videos

arXiv:2608.04939v1 Announce Type: new Abstract: Social media videos often communicate meanings that go beyond their visible actions, captions, or speech. A mundane clip may become humorous, ironic, or

Same Formulas, Different Semantics: Do Language Models Follow Modal Logic Specifications?

arXiv:2608.05097v1 Announce Type: new Abstract: Reasoning about necessity and possibility depends on assumptions about accessibility between worlds and about which objects exist at each one. The same

Seeking Physics in Diffusion Noise

arXiv:2603.14294v3 Announce Type: replace-cross Abstract: Do video diffusion models encode signals predictive of physical plausibility? We probe intermediate denoising representations of pretrained Di

researcharxiv-cs-ai

Adaptive Modality Reliability Diagnosis and Restoration for Robust Multimodal Intent Recognition

arXiv:2608.03475v1 Announce Type: cross Abstract: Multimodal intent recognition combines linguistic, acoustic, and visual evidence, but individual modalities may be noisy, missing, semantically confli

researcharxiv-cs-ai

ANCHOR-RE: An Agentic Neuro-Symbolic Framework for Grounded Biomedical Relation Extraction

arXiv:2608.03154v1 Announce Type: new Abstract: Biomedical relation extraction (BioRE) extracts structured knowledge from biomedical literature for applications such as knowledge base construction and

Balancing Efficiency and Efficacy: Training-Free Attention-Guided Switching Between Explicit and Latent Thoughts for MLLMs

arXiv:2608.03450v1 Announce Type: cross Abstract: Reasoning in Multimodal Large Language Models (MLLMs) requires both fine-grained visual perception and rigorous logical deduction. Explicit text-based

ChartAnno: Evaluating MLLMs for Chart Annotation Generation

arXiv:2608.03464v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have made significant progress in chart understanding, generation, and editing, but their ability to annotate e

Deferred Exposure of Future Trajectories for Verifiable Reasoning in Autonomous Driving VLMs

arXiv:2608.01755v2 Announce Type: replace Abstract: Recent Vision-Language-Action (VLA) models for autonomous driving (AD) increasingly utilize chain-of-thought (CoT) supervision to enhance the reason

safetyarxiv-cs-ai

Keep the Needle, Prune the Haystack: Defect-Preserving Token Pruning for Efficient Zero-Shot Anomaly Detection

arXiv:2608.03681v1 Announce Type: new Abstract: Zero-shot visual anomaly detection has achieved remarkable progress, with recent vision-only approaches further improving performance while simplifying

local-aiarxiv-cs-cv

LeanMem: Simple and Efficient Long-Term Memory for LLM Agents

arXiv:2608.03463v1 Announce Type: new Abstract: Long-term memory is essential for LLM-based agents to sustain interactions and reliably leverage distant history. However, existing memory systems typic

Hardware

LLM Serving in the Wild: An Empirical Study of Frameworks, Methods, and System Designs

arXiv:2608.03036v1 Announce Type: cross Abstract: Large Language Models (LLMs) are integrated into software systems and AI services, making efficient LLM serving a concern for software engineering. Se

hardwarearxiv-cs-ai

MultiCompose: Multi-Concept Personalized Composition with Per-Subject Attribute Binding

model-releasesarxiv-cs-cv

arXiv:2608.03708v1 Announce Type: new Abstract: Text-to-image diffusion models enable personalization of specific visual concepts from a small number of reference images. However, generating a single

NOMADD: Numerical Optimization of Models Adapting to Data Drift

model-releasesarxiv-cs-lg

arXiv:2608.02845v1 Announce Type: new Abstract: Tabular model performance degrades when feature distributions change over time or the relationship between features and outcome variables change over ti

PLAN: Parallel Liquid-Inspired Approximation Network for Efficient Representation Learning in Flexible Job Shop Scheduling

arXiv:2608.03041v1 Announce Type: cross Abstract: Deep reinforcement learning (DRL) approaches for flexible job shop scheduling (FJSP) heavily rely on attention-centric architectures to achieve state-

Self-Guided Adaptive Safety Alignment: Synthesizing and Internalizing Guidelines in Reasoning Models

arXiv:2511.21214v4 Announce Type: replace-cross Abstract: Explicit safety policies can improve reasoning-model safety, but their effective coverage may lag behind evolving jailbreak strategies. We stu

safetyarxiv-cs-ai

SparSEEty: Extracting Tokens from Sparsity-Exploiting LLM Serving Systems via Deterministic Side Channels

arXiv:2608.02995v1 Announce Type: cross Abstract: Modern large language models (LLMs) exhibit activation sparsity, wherein only a subset of their neurons is activated for given input tokens. Researche

researcharxiv-cs-ai

UniNav: A Unified World-Action Diffusion Model for Visual Navigation

arXiv:2608.03244v1 Announce Type: new Abstract: Image-goal visual navigation is a fundamental capability for embodied agents. Existing navigation policies efficiently predict waypoint trajectories but

When and Where to Look: Adaptive Visual Evidence Scheduling for Efficient Long Video Understanding

arXiv:2608.03918v1 Announce Type: cross Abstract: Efficient long-video understanding requires vision--language models (VLMs) to reason over a small number of frames selected as sparse visual evidence.

AdaMTP: An Adaptive Training Paradigm for Multi-Token Prediction

arXiv:2608.00434v1 Announce Type: new Abstract: Multi-Token Prediction (MTP) has emerged as an effective paradigm that augments a shared Large Language Model backbone with auxiliary heads, training th

Artificial Intelligence for the Characterization of Particles and Fibers by Optical Microscopy

arXiv:2608.00361v1 Announce Type: new Abstract: Optical microscopy of particle and fiber dispersions involves interpreting subtle visual cues influenced by specimen morphology, chemical composition, m

researcharxiv-cs-cv

Beyond Global Latents: Chunk-Based Sparse Grid VAE for Scalable 3D Modeling

arXiv:2608.02016v1 Announce Type: new Abstract: Sparse voxel grids preserve the spatial structure needed for detailed 3D reconstruction, but their memory still grows rapidly with resolution as active

local-aiarxiv-cs-cv

Coverage-Driven Adaptive Keyframe Selection for Video Understanding

arXiv:2608.00714v1 Announce Type: new Abstract: Recent advances in large vision-language models (LVLMs) have enabled long-video understanding and analysis. However, processing the large number of fram

tutorialsarxiv-cs-cv

DAVET: Denoising-Aware Visual Evidence Trajectory Allocation for Diffusion Vision-Language Models

arXiv:2608.01821v1 Announce Type: new Abstract: Diffusion vision-language models (dVLMs) iteratively denoise masked responses while conditioning each denoising step on visual evidence, making visual c

safetyarxiv-cs-cv

Device-First Feedback: Toward Mobile-Native LLM-Driven Neural Architecture Search

arXiv:2608.00078v1 Announce Type: new Abstract: Deploying convolutional neural networks generated by large language models (LLMs) on real mobile hardware requires more than GPU validation accuracy: IN

local-aiarxiv-cs-cv

DiffPrune: differentiable information throttling for token pruning in vision-language models

arXiv:2608.01985v1 Announce Type: new Abstract: Visual token pruning reduces the computational cost of Vision-Language Models (VLMs) by removing redundant visual tokens. The key is to learn a score th

tutorialsarxiv-cs-cv

Divergent large language model predictions from convergent representations in ambiguous word pairs

arXiv:2608.01816v1 Announce Type: new Abstract: In this work we investigate how decoder-only transformers resolve lexical ambiguity through layer-by-layer analysis of three models spanning three param

DS@GT ARC at MEDIQA-CORE-Task-1 2026: Trimodal Model Fusion with Task-Specific Gates for Brain Tumor Subtype Classification

model-releasesarxiv-cs-cv

arXiv:2608.00086v1 Announce Type: new Abstract: Brain tumor diagnosis is a time-sensitive process in which patients may wait weeks for a finalized pathology report. This problem motivates automated sy

Enhancing Large Language Model Reasoning with Reward Models: An Analytical Survey

arXiv:2510.01925v3 Announce Type: replace Abstract: Reward models (RMs) play a critical role in enhancing the reasoning performance of LLMs. For example, they can provide training signals to finetune

researcharxiv-cs-cl

EOVSAM: Efficient Open-Vocabulary Segmentation with SAM 3 in One Pass

arXiv:2608.02284v1 Announce Type: new Abstract: Open-vocabulary segmentation identifies and segments objects from arbitrary textual descriptions. SAM 3 supports noun-phrase-guided segmentation and ach

researcharxiv-cs-cv

From We to Me: Theory Informed Narrative Shift with Abductive Reasoning

arXiv:2603.03320v2 Announce Type: replace Abstract: Effective communication often relies on aligning a message with an audience's narrative and worldview. Narrative shift involves transforming text to