Skip to content

Paper Digest

Tags

🏷️ Tags¶

Browse digested papers by tag.

agent-harness¶

agentic-ai¶

agentic-framework¶

agentic-workflow¶

benchmark¶

chain-of-thought¶

clip¶

code-generation¶

contrastive-learning¶

credit-assignment¶

embodied-ai¶

evaluation¶

fine-tuning¶

foundation-model¶

grounding¶

grpo¶

gui-agent¶

hallucination¶

harness¶

human-in-the-loop¶

interpretability¶

A Comprehensive Information-Decomposition Analysis of Large Vision-Language Models
A Vision-Language Foundation Model for Zero-shot Clinical Collaboration and Automated Concept Discovery in Dermatology
ARCANE: A Multi-Agent Framework for Interpretable and Configurable Alignment
ARCANE: A Multi-Agent Framework for Interpretable and Configurable Alignment
ARIAL: An Agentic Framework for Document VQA with Precise Answer Localization
Abstract Counterfactuals for Language Model Agents
AgentTrace: Causal Graph Tracing for Root Cause Analysis in Deployed Multi-Agent Systems
Analysis-by-Proxy: Localization Signals in VLMs Operating as Condition Encoders
Ask Twice, Look Twice: Prompt Echoing Resolves the Question-First Paradox in Vision-Language Models
Attention-Guided Saliency Maps for Interpreting Visualization Literacy in VLMs
Attributes Should Come from Images, Not Class Names: Distribution-Conditioned Attribute Selection for Vision-Language Models
Bridging Vision and Language Concepts through Optimal Transport Semantic Flow
Concept-RuleNet: Grounded Multi-Agent Neurosymbolic Reasoning in Vision Language Models
Cross-Modal Redundancy and the Geometry of Vision–Language Embeddings
Cross-modal Associations in Vision and Language Models: Revisiting the Bouba-Kiki Effect
Decomposition of Concept-Level Rules in Visual Scenes
DiMaS: Distribution Matching for Steering Vision-Language-Action Models
Evaluating the Interpretability of Sparse Autoencoders with Concept Annotations
Explaining CLIP Zero-shot Predictions Through Concepts
Explaining Similarity in Vision-Language Encoders with Weighted Banzhaf Interactions
From Human Attention to Diagnosis: Semantic Patch-Level Integration of Vision-Language Models in Medical Imaging
GAM-Agent: Game-Theoretic and Uncertainty-Aware Collaboration for Complex Visual Reasoning
Hallucination Reduction with CASAL: Contrastive Activation Steering for Amortized Learning
Harnessing Textual Refusal Directions for Multimodal Safety
High-order Interactions Modeling for Interpretable Multi-Agent Q-Learning
LICORICE: Label-Efficient Concept-Based Interpretable Reinforcement Learning
LLM Agents Implement an NLG System from Scratch: Building Interpretable Rule-Based RDF-to-Text Generators
Label-Free Mitigation of Spurious Correlations in VLMs using Sparse Autoencoders
Language Models Can Explain Visual Features via Steering
Misaligned Roles, Misplaced Images: Structural Input Perturbations Expose Multimodal Alignment Blind Spots
Mitigating Object Hallucinations in LVLMs via Attention Imbalance Rectification
Mixture of Cognitive Experts in Large Vision-Language Models
One Head to Rule Them All: Amplifying LVLM Safety through a Single Critical Attention Head
Same Concept, Different Directions: Cross-Modal Feature Heterogeneity in Sparse Autoencoders
Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models
Sparse CLIP: Co-Optimizing Interpretability and Performance in Contrastive Learning
Structural Graph Probing of Vision-Language Models
Structured Reasoning for Fairness: A Multi-Agent Approach to Bias Detection in Textual Data
The Double Contingency Problem: AI Recursion and the Limits of Interspecies Understanding
Through the PRISM: Principle-Aware, Interpretable, and Multi-Scale Evaluation of Visual Designs
Utility Engineering: Analyzing and Controlling Emergent Value Systems in AIs
VISTA: A Vision and Intent-Aware Social Attention Framework for Multi-Agent Trajectory Prediction
ValuePilot: A Two-Phase Framework for Value-Driven Decision-Making
Vision Foundation Models for Computed Tomography
Vision Transformers Don't Need Trained Registers
Visual Thoughts: A Unified Perspective of Understanding Multimodal Chain-of-Thought
Visuals Lie, Consistency Speaks: Disentangling Spatial Attention from Reliability in Vision-Language Models
Visuals Lie, Consistency Speaks: Disentangling Spatial Attention from Reliability in Vision-Language Models

knowledge-distillation¶

knowledge-graph¶

llm-agent¶

llm-evaluation¶

llm-orchestration¶

lora¶

mcp¶

mechanistic-interpretability¶

medical-imaging¶

meta-harness¶

mllm¶

multi-agent¶

multimodal¶

multimodal-llm¶

multimodal-reasoning¶

orchestration¶

prompt-injection¶

rag¶

reasoning¶

red-teaming¶

reinforcement-learning¶

retrieval-augmented-generation¶

robot-manipulation¶

robustness¶

safety¶

security¶

software-engineering¶

spatial-reasoning¶

synthetic-data¶

tool-use¶

training-free¶

video-understanding¶

visual-grounding¶

visual-reasoning¶

vla¶

vlm¶

world-model¶

zero-shot¶