![[2501.01926] Cross-Modal Attention Calibration for LVLM Hallucination Mitigation](/static/img/curated/desks/afbytes-tech.w1200.webp)
Abstract page for arXiv paper 2501.01926: Cross-Modal Attention Calibration for LVLM Hallucination Mitigation
![[2501.04661] Beyond Memorization: Assessing Semantic Generalization in Large Language Models Using Phrasal Constructions](/static/img/curated/desks/afbytes-tech.w400.webp)
Abstract page for arXiv paper 2501.04661: Beyond Memorization: Assessing Semantic Generalization in Large Language Models Using Phrasal Constructions
Abstract page for arXiv paper 2502.12119: PRISM: Self-Pruning Intrinsic Selection Method for Training-Free Multimodal Data Selection
Abstract page for arXiv paper 2502.15224: Auto-Discovery-Bench: Diagnosing Structured State Tracking in Oracle-Guided Discovery
Abstract page for arXiv paper 2503.05846: EMCEE: Improving Multilingual Capability of LLMs via Bridging Knowledge and Reasoning with Extracted Synthetic Multili...
Abstract page for arXiv paper 2505.22934: Unraveling LoRA Interference: Orthogonal Subspaces for Robust Model Merging
Abstract page for arXiv paper 2506.08255: SHIELD: Secure Hypernetworks for Incremental Expansion Learning Defense
Abstract page for arXiv paper 2605.30561: VLM3: Vision Language Models Are Native 3D Learners
Abstract page for arXiv paper 2605.30571: Memory-Bound but Not Bandwidth-Limited: The Physical AI Inference Gap in Batch-1 LLM Decode
Abstract page for arXiv paper 2605.30581: Prior Availability in Industrial Visual Sim-to-Real: A Review of CAD-Guided and CAD-Unavailable Regimes
Abstract page for arXiv paper 2605.30585: Benchmarking Machine Learning Uncertainty Quantification Methodologies for Predicting Turbine Gas Temperature Degradat...
Abstract page for arXiv paper 2605.30589: ImmigrationQA: A Source-Grounded Dataset and Small-Model Adaptation for U.S. Immigration Law
Abstract page for arXiv paper 2605.30590: Counterfactual Evaluation Reveals Hidden Capability Profiles in Clinical LLMs and Agents
Abstract page for arXiv paper 2605.30593: Scientific Machine Learning for Engine Health Management and Remaining Useful Life Prediction
Abstract page for arXiv paper 2605.30604: An Organization-Scoped LLM Agent Runtime Architecture for Regulated Cybersecurity Operations
Abstract page for arXiv paper 2605.30611: Crafter: A Multi-Agent Harness for Editable Scientific Figure Generation from Diverse Inputs
Abstract page for arXiv paper 2605.30628: The Architecture of Errors: From Universal Impossibility to Patch-Local LLM Reliability
Abstract page for arXiv paper 2605.30381: When LLMs Learn to Be Consistently Wrong: A Multi-Model Study of Linear Representations of Synthetic Deception
Abstract page for arXiv paper 2605.30675: Human-Alignment, Calibration, and Activation Patterns in Large Language Model Uncertainty
Abstract page for arXiv paper 2605.30677: Investigating Detection and Obfuscation of Prompt Injection Attacks Against Software Reverse Engineering AI Agents
Abstract page for arXiv paper 2605.30685: How Early Adopters Used Generative AI Worldwide: Variation by Country Income and Language
Abstract page for arXiv paper 2605.30689: ConTrans: Learning Text-enhanced Local-global Temporal Representations for Zero-shot Temporal Action Localization
Abstract page for arXiv paper 2605.30698: Seeing Before Agreeing: Aligning Multi-Agent Consensus with Visual Evidence
Abstract page for arXiv paper 2605.30716: Simple Token-Efficient Vision-Language Model for Case-level Pathology Synoptic Report Generation
Abstract page for arXiv paper 2605.30719: When are LLMs Sufficient Policy Optimizers for Sequential RL Tasks?
Abstract page for arXiv paper 2605.30736: OrcaRouter: A Production-Oriented LLM Router with Hybrid Offline-Online Learning
Abstract page for arXiv paper 2605.30740: GSAM: A Generalizable and Safe Robotic Framework for Articulated Object Manipulation
Abstract page for arXiv paper 2605.30911: What Makes LVLMs Hallucinate Less? Unveiling the Architectural Factors Behind Hallucination Robustness
Abstract page for arXiv paper 2605.30913: Toxic HallucinAItions: Perturbing Prompts and Tracing LLM Circuits
Abstract page for arXiv paper 2605.30919: De-attribute to Forget for LLM Unlearning
Abstract page for arXiv paper 2605.30930: TUX: Measuring Human--AI Tacit Understanding
Abstract page for arXiv paper 2605.30934: Do Large Language Models Encode Institutional Experience? Evidence from Cross-Linguistic Moral Reasoning Under Ambigui...
Abstract page for arXiv paper 2605.30963: AMix-2: Establishing Protein as a Native Modality in Large Language Models
Abstract page for arXiv paper 2605.30965: ImmersiveTTS: Environment-Aware Text-to-Speech with Multimodal Diffusion Transformer and Domain-Specific Representatio...
Abstract page for arXiv paper 2605.30966: Reading Between the Citations: A Typed Claim Network for Scientific Literature
Abstract page for arXiv paper 2605.30968: Variational Adapter for Cross-modal Similarity Representation
Abstract page for arXiv paper 2605.30984: Generating Reports or Repeating Templates? Measuring and Mitigating Template Collapse in 3D CT Report Generation
Abstract page for arXiv paper 2605.31289: The Terminal Representation in Reinforcement Learning
Abstract page for arXiv paper 2605.31295: Latent Space Disentanglement via Activation Steering for Interpretable Attribute Control in Symbolic Music Generation
Abstract page for arXiv paper 2605.31324: Inconsistency-Aware Minimization: Improving Generalization with Unlabeled Data
Abstract page for arXiv paper 2605.31330: Social welfare optimisation under institutional reward and punishment
Abstract page for arXiv paper 2605.31340: Appropriateness of Empathy in AI: A Signal-Cost Perspective
Abstract page for arXiv paper 2605.31349: FBHM: Functional Benchmarking and Steering of VLMs for Hateful Meme Detection
Abstract page for arXiv paper 2605.31360: dashi: A Python library for Dataset Shift Characterization to Support Trustworthy AI Development and Deployment
Abstract page for arXiv paper 2605.31361: Dreaming Of Others: Latent Teammate Modeling In World Models For Multi-Agent Reinforcement Learning
Abstract page for arXiv paper 2605.31251: ERGeoBench:A Comprehensive Benchmark for Embodied Reasoning and Geo-localization in Multimodal Large Language Models
Abstract page for arXiv paper 2605.31393: Target-Side Paraphrase Augmentation for Sign Language Translation with Large Language Models