The Archive

Subject-conditioned causal neural surrogate for pediatric cerebral palsy rehabilitation using OpenSim-derived parameters.

Shallow Recurrent Decoder for state estimation in tokamak liquid metal blankets via reduced-order modeling.

EvoUndo framework ensures LLM agent self-modifications are reversible across counterfactual states; identifies 197 safe mutations from 600 tasks.

Optimal testing strategy to recover honest results from adversarial test takers including AI-aided cheating.

GRACE gradient-guided coreset selection constructs forget/retain sets for LLM unlearning without pre-specification.

Framework propagates clinical guideline KG construction-time quality signals into medical QA evidence selection.

AGENT-O ontology framework standardizes semantic representation and governance reporting for healthcare AI agents across 279 scientific publications.

Cross-spectral dense correspondence method enables multimodal medical imaging fusion across disparate wavelength ranges.

Fractional Power Encoding extends hyperdimensional computing with Hadamard product binding for shift-equivariant sequence representations.

BanglaMed-QA introduces medical question-answering system for low-resource Bangla language with 4,493 QA pairs across 506 diseases.

Study measures failure correlation between stacked LLM defenses using Adversary Access-Tier Model, finding defense layers often fail on same inputs.

MAIL framework applies memory-augmented LLMs to hypothesis generation in chemistry by adaptively navigating chemical literature.

AIM method enables MLLM unlearning for identity-specific data deletion without retain images by separating identity and visual-perception hidden states.

Scaling laws study characterizes optimal learning rate and batch size evolution with model capacity for OpenEuroLLM pretraining.

VISTA improves on-policy self-distillation by adaptively aligning student and teacher distributions using verifier feedback for reasoning tasks.

PanelShield framework enables verifiable closed-loop safety planning for industrial robotic panel operation with constraint satisfaction and violation repair.

MaCoPlanner compiles equipment manuals into typed representations for LLM-assisted robotic task planning with symbolic safety verification.

Memristive-friendly reservoir computing replaces dense matrices with structured orthogonal operators for neuromorphic hardware deployment.

Formalizes KV cache eviction via probabilistic reasoning to improve LLM inference throughput with quality-preserving heuristics.

Dense embedding models for semantic search incorporate dual assessment of topic-relevance and stance in computational argumentation.

A federal judge ruled the Trump administration illegally labeled Anthropic a supply chain risk, handing the AI company a victory as its second Pentagon lawsuit continues in Washington.

LoopArena benchmarks models as runtime loop controllers for coding agents, isolating loop guidance quality from agent capability.

PASK optimizes KV persistence for structured generation by tracking parser transitions to preserve schema-critical decoding signals.

Proves Colombo's 1928 determinant conjecture for difference-power matrices using algebraic methods.

Transfer learning framework for unified urban mobility forecasting across heterogeneous transportation modes with limited data.

RECAST performs test-time adaptation on streaming biosignals by sampling recent, context-aligned, reliable observations for online personalization.

Real-time semantic navigation framework for UAV object-goal search using LLMs to reason over spatial context and prioritize regions.

Sandhya Devanathan will oversee some OpenAI operations across Southeast Asia and Australia in her new role.

Microsoft postpones Teams Facilitator AI bot launch by 2 months for refinement.

OpenAI claims path to AGI by end-2026; timeline assertion lacks technical details or verification criteria.