All Articles
17208 articles total
Knowledge Graphs and Explainable AI as Complementary Resources for Urban Mining
Explores the integration of Knowledge Graphs and Explainable AI (XAI) to improve the defensibility of AI-supported decisions in urban mining audits.
ARCANA: A Reflective Multi-Agent Program Synthesis Framework for ARC-AGI-2 Reasoning
ARCANA is a multi-agent framework designed to solve ARC-AGI-2 tasks using iterative perception, hypothesis generation, and symbolic execution.
Neuro-Agentic Control: A Deep Learning-based LLM-Powered Agentic AI Framework for Controlling Security Controls
Neuro-Agentic Control combines LLM planners with Time-Series Foundation Models (TimesFM) to create a safe, physics-grounded autonomous defense for industrial IoT.
L-MAD: A Systematic Evaluation of Multi-Agent Debate Structures in Legal Reasoning
The L-MAD framework evaluates multi-agent debate structures for legal reasoning, finding a trade-off between agent population and discussion rounds.
MedRealMM: A Real-World Multimodal Benchmark for Chinese Online Medical Consultation
MedRealMM is a new real-world multimodal benchmark for Chinese online medical consultation to evaluate the clinical quality and safety of LLMs.
KV-PRM: Efficient Process Reward Modeling via KV-Cache Transfer for Multi-Agent Test-Time Scaling
KV-PRM introduces an efficient process reward model that reads KV caches directly, reducing scoring costs from O(L^2) to O(L) for test-time scaling.
Scoped Verification for Reliable Long-Horizon Agentic Context Evolution under Distribution Shift
GRACE proposes a graph-regularized agentic context evolution system to maintain and verify persistent system instructions for LLM agents over long horizons.
Toward Auditable AI Scientists: A Hypothesis Evolution Protocol for LLM Agents
The Hypothesis Evolution Protocol (HEP) provides an auditable harness for LLM agents to generate, test, and evolve scientific hypotheses.
OpenProver: Agentic and Interactive Theorem Proving with Lean 4
OpenProver is an open-source system for LLM-driven automated theorem proving that integrates with Lean 4 for formal verification.
LongMedBench: Benchmarking Medical Agents for Long-Horizon Clinical Decision-Making
LongMedBench is a real-world EHR-based benchmark designed to test the long-horizon clinical decision-making capabilities of medical agents.
Communication-Efficient Digital-Twin Coordination for Heterogeneous LLM Embodied Agents over Computing Power Networks
LDT-Coord uses lightweight digital twins to coordinate heterogeneous LLM embodied agents, reducing communication overhead by 70x.
Count Binface
Discussion regarding Count Binface, a satirical political candidate in the UK.
Are you telling me a readonly property is wrecking my performance?
A technical discussion on how readonly properties can unexpectedly degrade performance in certain software environments.
Interval Certifications for Multilayered Perceptrons via Lattice Traversal
Presents a theoretical framework for AI safety and adversarial robustness using lattice traversal and introduces the ParallelepipedoNN system.
CogniConsole: Externalizing Inference-Time Control as a Formal Abstraction for Reliable LLM Interactions
Introduces CogniConsole, an architectural abstraction that externalizes inference-time control to improve LLM reliability and reduce variance.
GATS: Graph-Augmented Tree Search with Layered World Models for Efficient Agent Planning
Proposes GATS, a planning framework that uses a layered world model to eliminate LLM calls during inference for agent planning.
Long-Horizon-Terminal-Bench: Testing the Limits of Agents on Long-Horizon Terminal Tasks with Dense Reward-Based Grading
Introduces Long-Horizon-Terminal-Bench, a demanding benchmark for evaluating AI agents on long-term terminal tasks with dense rewards.
A Formalization of the Mean-Field Derivation of the Vlasov Equation: AI-Assisted Lean Formalization as a Strategy Game
Demonstrates an AI-assisted strategy for formalizing the Vlasov Equation in Lean 4, treating the process as a formalization game.
Quadrupling code performance with a "useless" if
An exploration of how a seemingly unnecessary conditional check can significantly optimize code performance through branch prediction or memory alignment.
Why are hard drives so expense
A user expresses frustration over the rising cost of hard drives, attributing the trend to high demand from AI and data centers.
