A new comprehensive survey indexes 120 sign-language datasets across 35 sign languages, identifying key challenges such as modality imbalance and annotation inconsistency. The authors introduce a standardized 24-field datasheet and a public GitHub repository to promote reproducible research in sign-language technologies.

Proposes LUCID, a hallucination detection method that leverages LLM attention scores, KG semantics, and structural information via graph neural networks, achieving state-of-the-art performance on nine datasets.

Researchers propose Causal Attribution Pruning (CAP), a training-free method that identifies critical attention heads in LLMs by measuring their causal impact on reasoning tasks. CAP achieves up to 61% relative accuracy improvement over Wanda at 20% sparsity on ARC-Challenge, and consistently outperforms at moderate sparsity levels.

This paper reveals that query position critically affects in-context learning in diffusion LLMs due to a recency effect in attention. The authors propose Average Confidence metric and Auto-ICL, a training-free strategy to optimize query placement, achieving robust performance across tasks.

A new study introduces an error taxonomy for LLMs in hardware design, revealing a ceiling on VerilogEval at 90.8% pass rate due to unsolvable functional errors. Alignment techniques only teach compilation, not reasoning.

This study fine-tunes seven large language models on Arabic and evaluates zero-shot reading comprehension on Semitic and non-Semitic languages. No evidence of Semitic-specific transfer was found; improvements are due to task-format alignment, not linguistic knowledge.

This study investigates using Google's Gemini and Gemma LLMs to automate detection of EQ-5D studies in PubMed from abstracts. A multi-phase framework with few-shot prompting, weight ensembling, and soft stacking meta-classifier is proposed. Evaluating nine LLMs on expert-labeled data, the weighted ensemble of gemini-2.5-pro, gemma-3-12b, and gemma-3-27b achieved 0.74 weighted F1-score and accuracy, outperforming individual models. Ensembling improved precision-recall balance, and soft stacking enhanced reliability and interpretability. Results suggest ensemble LLMs are a reliable, scalable approach for automating biomedical screening.

Ventricular tachycardia is a life-threatening arrhythmia. Pace-mapping identifies ablation targets. cAPM uses continual learning to transfer knowledge across VTs, reducing pacing sites needed. In silico, it achieved 81% localization accuracy with 4.5 sites vs. 38% with 13.7 for prior methods.

Self-play reinforcement learning can train driving policies without human data but often learns alien behaviors. A new method uses just 30 minutes of human demonstrations as a regularizer on top of a minimal safe reward, achieving human-compatible policies with 2500x less data than imitation learning, training in 15 hours on a single consumer GPU.

Proposes multivariate zero-inflated Gaussian (ZIG) distributions as sampling laws for EDAs, enabling joint optimization of sparsity patterns and active values without hand-crafted operators. The latent parameters are identifiable, and practical amortized inversion estimators are provided. On the Lunar Lander benchmark, ZIG-EDA outperforms dense Gaussian EDA, hand-crafted sparse EA, and ad-hoc sparse EDA in convergence speed and final returns while using few active parameters.

Building on a two-parameter Weibull framework, this paper analyzes why the Weibull weight-scale parameter λ grows, overshoots, and relaxes during AdamW training. A three-force decomposition (alignment, injection, decay) is derived, showing alignment dominates the rise phase (88-94%). Near saturation, alignment and decay balance. A spline displacement method recovers alignment from sparse checkpoints with ~92-94% accuracy. Peak λ varies with training-data coherence, suggesting a data-dependent component.

Information lattice learning (ILL) learns interpretable rules of a signal by alternately projecting the signal onto a partition lattice and lifting selected rules back. When the signal is a probability mass function, the probabilistic rules learned by ILL admit a natural probabilistic graphical model (PGM) interpretation. A partition induces a deterministic quotient variable, and a rule is the marginal law of that variable. A rule set is a collection of marginal constraints over interpretable abstractions. General lifting is the feasible family of all joint distributions satisfying constraints, while special lifting chooses a maximum-ignorance reconstruction via an L2 uniformity principle related to maximum entropy. Under Shannon-entropy lifting, constraints yield a log-linear factor graph. The information lattice itself is not a Bayesian network; its edges encode refinement/coarsening, not conditional dependence. Thus ILL is best viewed as structure learning for interpretable constraint-based factor graphs over quotient variables.

This paper analyzes Med-DDPM's performance across three NVIDIA GPU generations, identifies inefficiencies, and demonstrates TF32 Tensor Core and channels-last optimizations achieving up to 100x improvement without quality loss.

A new paper introduces SPSD (Sentiment Preserving Semantic Distillation), an edge-based pipeline that compresses user prompts by removing social scaffolding before sending them to a cloud LLM, reducing input tokens by an average of 99.9 per call while maintaining response quality within a non-inferiority margin. The approach aims to cut cloud energy costs.

A novel framework called Gated Uncertainty-Aware Routing for Distillation (Guard) is proposed to address distributional misalignment and computational cost of Time-Series Foundation Models in scientific applications. It uses a Contextual Router and Uncertainty-Gated Temperature to distill knowledge from multiple pretrained models into lightweight forecasters. Evaluated on four climate-critical domains, Guard significantly reduces RMSE compared to a fixed-weight multi-teacher baseline, and misaligned teachers outperform globally superior models on 28.5% of hardest instances.

This paper introduces 'computational identifiability', a framework that replaces theoretical asymptotic assumptions with a finite computational search procedure for empirical estimators. Experiments show it works for small samples, ambiguous graphs, mixed data, and counterfactuals.

This paper proposes a prompt-based uncertainty decomposition method that separates action confidence from request uncertainty, enabling LLM agents to ask for clarification when task specifications are ambiguous. The authors introduce two new benchmarks with 50% underspecified tasks and evaluate against existing methods across five LLMs, showing significant F1 improvements.

This paper introduces the Integral Transform Network (ITNet), a unified architecture that generalizes convolution, self-attention, and recurrence through a learnable integral kernel. ITNet matches or exceeds specialized models on multiple benchmarks.

A new method enables LLMs to self-align with ethics using a conscience step and Direct Preference Optimization, without external judges. The technique counters emergent misalignment in scenarios like code hacking.

The paper proposes REVEAL++, a continuous formulation of phenotypic grouping in contrastive learning for vision-language alignment using retinal images and clinical risk narratives to predict Alzheimer's disease risk. It replaces hard group assignments with differentiable weights, enabling graded supervision and end-to-end learning. Evaluated on UK Biobank, it outperforms discrete methods.

A study comparing Qwen 2.5 7B and XGBoost on clinical prediction reveals that LLM verbalized confidence is epistemically vacuous, an inverse difficulty effect exists, few-shot and SHAP interventions improve accuracy, and a cross-model calibrator reduces calibration error.

DeXposure-Claw is a forecast-grounded agentic supervision system for decentralized finance risk, addressing the over-reading and high-stakes false alarms of general-purpose LLM agents. It uses a graph time-series foundation model to predict exposure networks, deterministic monitors and stress scenarios to generate alerts, and data-health gates to constrain escalation. DeXposure-Bench evaluates system decisions on six axes, including a regulator-aligned false-intervention rate. Experiments on five years of real data validate the system.

This paper proposes a dynamical system model for multi-agent LLM deliberation, where each agent has a hidden internal belief (anchor) that continuously pulls its opinion. The model explains a behavior forbidden by classical consensus rules: an agent's confidence can exceed the convex hull of initial beliefs. Experiments on three open-weight model families show that all anchors have similar influence but differ in location, and only when the anchor is far from initial opinions does deliberation escape the hull.

This paper presents a systematic experimental analysis of eight state-of-the-art diffusion language models (DLMs) across eight benchmarks spanning reasoning, coding, translation, knowledge, and structured problem solving, considering both generation quality and computational efficiency. It examines the impact of inference-time factors such as denoising steps, context length, block size, and parallel unmasking strategies, and finds that DLM behavior is strongly influenced by generation-time design choices, leading to distinct trade-offs between performance and efficiency. The study provides practical insights into the capabilities and deployment characteristics of contemporary DLMs.

A new study introduces a human-in-the-loop pipeline to measure how well undergraduate computer science programs align with curricular guidelines. Applied to CS2013 and CS2023, it found near-constant coverage (~50%) but a drop in cognitive depth achievement from 95% to 76%, reflecting raised expectations in the newer standard. Persistent gaps in parallel computing, programming languages, and systems fundamentals were identified.

A new paper proposes AgenticRei, a deontic policy framework to govern LLM-driven autonomous agents, addressing obligations, dispensations, and policy conflicts beyond current access-control engines.

Sqim is a new tool that allows developers to install iOS app builds directly onto iPhones without needing VPN or Tailscale. It installs via Homebrew, integrates with coding agents like Codex and Claude Code, and simplifies the iOS deployment process.

The author recounts building a local translation pipeline for Korean to English using a loop of planning, execution, critique, and repair, but ultimately failing to improve quality and waiting for better models.

We implement an end-to-end workflow for Salesforce CodeGen, loaded from Hugging Face. We move past basic inference by adding function extraction, syntax checking, static safety checks, and unit-test validation. We rerank best-of-N candidates, compose multi-turn program synthesis, and experiment with prompt styles. We finish by visualizing a mini benchmark and exporting the generated artifacts as reusable files.

pumaDB is a lightweight hosted memory layer for AI agents, offering shared memory via MCP or server-side API to store project context, research notes, transcripts, and more, without managing databases or vector stores.