Search: “models”

Causal Evidence for Attention Head Imbalance in Modality Conflict Hallucination

arXiv:2605.19250v1 Announce Type: new Abstract: Modality-conflict hallucination occurs when multimodal large language models (MLLMs) prioritize erroneous textual premises over contradictory visual evi

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

Compositional Literary Primitives in Instruction-Tuned LLMs: Cross-Architectural SAE Features for Self, Style, and Affect

DGX agent

arXiv:2605.18808v1 Announce Type: cross Abstract: We characterize a compositional architecture of literary primitives in two instruction-tuned large language models (Llama 3.1 8B-Instruct and Gemma 2

model-releasesarxiv-cs-ai

20 May 2026

DecisionBench: A Benchmark for Emergent Delegation in Long-Horizon Agentic Workflows

arXiv:2605.19099v1 Announce Type: new Abstract: We introduce DecisionBench, a benchmark substrate for emergent delegation in long-horizon agentic workflows. The substrate fixes a task suite (GAIA, tau

Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth

arXiv:2605.19797v1 Announce Type: new Abstract: Monocular depth estimation has improved significantly in recent years, driven by increasingly powerful models and large-scale training data. Predicted d

Local Ai

Descriptive versus Regulatory Uncertainty in Bounded Predictive Systems

arXiv:2605.18909v1 Announce Type: new Abstract: Any system that models the world under finite representational capacity must compress; any compression entails a prior; and the prior is the system's bi

local-aiarxiv-cs-lg

Differential-Integral Neural Operator for Long-Term Turbulence Forecasting

arXiv:2509.21196v3 Announce Type: replace-cross Abstract: Accurately forecasting the long-term evolution of turbulence represents a grand challenge in scientific computing and is crucial for applicati

DMN: A Compositional Framework for Jailbreaking Multimodal LLMs with Multi-Image Inputs

arXiv:2605.18915v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) are vulnerable to jailbreak attacks, which can elicit harmful responses from MLLMs. Many MLLMs support multi-

DocQT: Improving Document Forgery Localization Robustness via Diverse JPEG Quantization Tables

arXiv:2605.19688v1 Announce Type: new Abstract: Document manipulation localization models achieve strong performance on public benchmarks yet fail to generalize to operational document workflows. We i

EventPrune: Cascaded Event-Assisted Token Pruning for Efficient First-Person Dynamic Spatial Reasoning

arXiv:2605.19506v1 Announce Type: new Abstract: First-person dynamic spatial reasoning requires models to track continuous motion and precise geometric structure, but the quadratic attention cost of T

Feature-Space Smoothing: Certified Robustness of Deep Representations

arXiv:2601.16200v3 Announce Type: replace-cross Abstract: Modern deep learning models exhibit strong capabilities across diverse applications, yet remain vulnerable to malicious inputs that induce err

safetyarxiv-cs-cv

Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents

arXiv:2605.19604v1 Announce Type: new Abstract: Large Language Model (LLM) agents increasingly act inside real workspaces, where tools and skills determine whether model reasoning becomes reliable act

safetyarxiv-cs-ai

Generalization Bounds of Surrogate Policies for Combinatorial Optimization Problems

arXiv:2407.17200v3 Announce Type: replace-cross Abstract: Many real-world decision problems require solving, again and again, combinatorial optimization instances drawn from a common distribution. A r

GoTTA be Diverse: Rethinking Memory Policies for Test-Time Adaptation

arXiv:2605.19890v1 Announce Type: new Abstract: Test-time adaptation (TTA) enables a pre-trained model to adapt online to an unlabeled test stream under distribution shift. While most TTA research foc

HAVEN: Hierarchically Aligned Multimodal Benchmark for Unified Video Understanding

arXiv:2605.19223v1 Announce Type: new Abstract: While Multimodal Large Language Models (MLLMs) exhibit strong performance on standard video tasks, their ability to faithfully summarize and reason over

Hybrid-LoRA: Bridging Full Fine-Tuning and Low-Rank Adaptation for Post-Training

arXiv:2605.18822v1 Announce Type: cross Abstract: Post-training has become essential for adapting large language models (LLMs) to complex downstream behaviors, including instruction following, prefere

Less Back-and-Forth: A Comparative Study of Structured Prompting

arXiv:2605.20149v1 Announce Type: cross Abstract: Large language models (LLMs) are widely used for open-ended tasks, but underspecified prompts can lead to low-quality answers and additional interacti

Library Hallucinations in LLM-Generated Code: A Risk Analysis Grounded in Developer Queries

arXiv:2509.22202v3 Announce Type: replace-cross Abstract: Large language models (LLMs) now play a central role in code generation, yet they continue to hallucinate, frequently inventing non-existent l

Research

Lossless Anti-Distillation Sampling

arXiv:2605.18829v1 Announce Type: new Abstract: Frontier commercial generative models face a growing threat from distillation, whereby a distiller harvests generated responses and trains a competing m

researcharxiv-cs-lg

MIRO: MultI-Reward cOnditioned pretraining improves T2I quality and efficiency

arXiv:2510.25897v2 Announce Type: replace Abstract: The default paradigm of post-training text-to-image generators includes post-hoc selection of generated images, and subsequent training with one rew

MVI-Bench: A Comprehensive Benchmark for Evaluating Robustness to Misleading Visual Inputs in LVLMs

arXiv:2511.14159v2 Announce Type: replace Abstract: Evaluating the robustness of Large Vision-Language Models (LVLMs) is essential for their continued development and responsible deployment in real-wo

Next-Acceleration-Scale Prediction for Autoregressive MRI Reconstruction

arXiv:2605.19354v1 Announce Type: cross Abstract: MRI reconstruction is an inherently ill-posed inverse problem, since incomplete measurements admit many plausible solutions. This ambiguity becomes mo

Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design

arXiv:2605.19717v1 Announce Type: new Abstract: Large Language Models (LLMs) can generate Computer-Aided Design (CAD), yet lack physical comprehension required for reliable engineering design. Instead

Position: Uncertainty Quantification in LLMs is Just Unsupervised Clustering

arXiv:2605.19220v1 Announce Type: cross Abstract: Uncertainty Quantification (UQ) is widely regarded as the primary safeguard for deploying Large Language Models (LLMs) in high-stakes domains. However

safetyarxiv-cs-ai

PromptRad: Knowledge-Enhanced Multi-Label Prompt-Tuning for Low-Resource Radiology Report Labeling

arXiv:2605.20052v1 Announce Type: cross Abstract: Automatic report labeling facilitates the identification of clinical findings from unstructured text and enables large-scale annotation for medical im

RECIPE: Procedural Planning via Grounding in Instructional Video

arXiv:2605.19976v1 Announce Type: new Abstract: Visual planning asks a model to generate the remaining steps of a procedure in natural language given a partial video context and a goal. Progress on th

Retrieval-Augmented Generation for Natural Language Processing: A Survey

arXiv:2407.13193v4 Announce Type: replace Abstract: Large language models (LLMs) have achieved strong empirical performance in various fields, benefiting from their huge amount of parameters that stor

Smooth Piecewise Cutting for Neural Operator to Handle Discontinuities and Sharp Transitions

arXiv:2605.19823v1 Announce Type: cross Abstract: Neural operators have achieved strong performance in learning solution operators of partial differential equations (PDEs), but their inherently contin

Applications

Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligence

arXiv:2505.23747v2 Announce Type: replace-cross Abstract: Recent advancements in Multimodal Large Language Models (MLLMs) have significantly enhanced performance on 2D visual tasks. However, improving

applicationsarxiv-cs-ai

Tail Annealing for Heavy-Tailed Flow Matching

arXiv:2605.20068v1 Announce Type: cross Abstract: Standard generative models struggle with heavy-tailed data: Lipschitz architectures cannot produce power-law tails from Gaussian noise, and interpolat

Taming the Thinker: Conditional Entropy Shaping for Adaptive LLM Reasoning

arXiv:2605.19358v1 Announce Type: new Abstract: Entropy-based deep reasoning has emerged as a promising direction for improving the reasoning capabilities of Large Language Models (LLMs), but existing

TEMPO: Temporal Enforcement via Mode-Separated Policy Optimization for Trustworthy LLM Backtesting

arXiv:2605.18843v1 Announce Type: new Abstract: Backtesting large language models on historical events requires reasoning exclusively from information available before a specified cutoff date. Yet mod

safetyarxiv-cs-lg

The Annotation Scarcity Paradox in Low-Resource NLP Evaluation: A Decade of Acceleration and Emerging Constraints

arXiv:2605.19066v1 Announce Type: new Abstract: Over the past decade, low-resource natural language processing (NLP) has experienced explosive growth, propelled by cross-lingual transfer, massively mu

The Silent Hyperparameter: Quantifying the Impact of Inference Backends on LLM Reproducibility

arXiv:2605.19537v1 Announce Type: new Abstract: Progress in LLMs is increasingly measured through standardized benchmarks, where state-of-the-art improvements are often separated by fractions of a per

Towards Consistent Detection of Cognitive Distortions: LLM-Based Annotation and Dataset-Agnostic Evaluation

arXiv:2511.01482v2 Announce Type: replace Abstract: Text-based automated Cognitive Distortion detection is a challenging task due to its subjective nature, with low agreement scores observed even amon

TwinRouterBench: Fast Static and Live Dynamic Evaluation for Realistic Agentic LLM Routing

arXiv:2605.18859v1 Announce Type: cross Abstract: LLM routing matters most in long-horizon applications such as coding agents, deep research systems, and computer-use agents, where a single user reque

What Do Evolutionary Coding Agents Evolve?

arXiv:2605.20086v1 Announce Type: cross Abstract: Recent work pairs LLMs with evolutionary search to iteratively generate, modify, and select code using task-specific feedback. These systems have prod

Whispers of Wealth: Red-Teaming Google's Agent Payments Protocol via Prompt Injection

arXiv:2601.22569v2 Announce Type: replace-cross Abstract: Large language model (LLM) based agents are increasingly used to automate financial transactions, yet their reliance on contextual reasoning e

A Feature-Driven Framework for Software Fault Prediction

arXiv:2605.17611v1 Announce Type: cross Abstract: Software fault prediction (SFP) is a critical task in software engineering, enabling early identification of faults in modules to improve software qua

A Machine With Human-Like Memory Systems

arXiv:2204.01611v3 Announce Type: replace Abstract: Inspired by the cognitive science theory, we explicitly model an agent with both semantic and episodic memory systems, and show that it is better th

Research

A Theory of Training Profit-Optimal LLMs

arXiv:2605.16430v1 Announce Type: cross Abstract: Scaling LLMs requires tremendous computational resources, and recent advances in AI have gone hand in hand with massive amounts of capital expenditure

researcharxiv-cs-ai

Local Ai

AdaptiveLoad: Towards Efficient Video Diffusion Transformer Training

arXiv:2605.17923v1 Announce Type: cross Abstract: In video generation models, particularly world models, training large-scale video diffusion Transformers (such as DiT and MMDiT) poses significant com

local-aiarxiv-cs-ai

Advancing Narrative Long Video Generation via Training-Free Identity-Aware Memory

arXiv:2605.18733v1 Announce Type: new Abstract: Autoregressive video generation has improved rapidly in visual fidelity and interactivity, but it still suffers from long-term inconsistency and memory

Agents

AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent

arXiv:2602.03955v2 Announce Type: replace Abstract: While large language model (LLM) multi-agent systems achieve superior reasoning performance through iterative debate, practical deployment is limite

agentsarxiv-cs-ai

AgentSteerTTS: A Multi-Agent Closed-Loop Framework for Composite-Instruction Text-to-Speech

arXiv:2605.17583v1 Announce Type: new Abstract: While existing text-to-speech (TTS) models exhibit high expressiveness, fine-grained control over composite instructions remains challenging due to the

AgentWall: A Runtime Safety Layer for Local AI Agents

arXiv:2605.16265v1 Announce Type: new Abstract: The safety of autonomous AI agents is increasingly recognized as a critical open problem. As agents transition from passive text generators to active ac

AuthorMix: Modular Authorship Style Transfer via Layer-wise Adapter Mixing

arXiv:2603.23069v2 Announce Type: replace-cross Abstract: The task of authorship style transfer involves rewriting text in the style of a target author while preserving the meaning of the original tex

Barriers for Learning in an Evolving World: Mathematical Understanding of Loss of Plasticity

arXiv:2510.00304v3 Announce Type: replace-cross Abstract: Deep learning models excel in stationary data but struggle in non-stationary environments due to a phenomenon known as loss of plasticity (LoP

Beyond Accuracy: Decomposing the Reasoning Efficiency of LLMs

arXiv:2602.09805v2 Announce Type: replace-cross Abstract: As reasoning LLMs increasingly trade tokens for accuracy through deliberation, search, and self-correction, a single accuracy score can no lon