A new study from arXiv argues that Equalized Odds is a more reliable fairness criterion than Demographic Parity for guiding bias correction in AI classification. The researchers introduce FairDream, a…
Researchers introduce TOM-GS, an editable video representation that uses regular 3D Gaussians with a continuous temporal opacity formulation, avoiding complex deformations. The method outperforms prio…
A study using EEG signals and a custom convolutional neural network (CNN) predicted the outcome of repetitive transcranial magnetic stimulation (rTMS) therapy for major depressive disorder (MDD) with …
A new preprint on arXiv (2607.22748v1) introduces Accessibility Plasticity, a principle of adaptive computation in which systems adapt by reorganizing which existing computations can interact, rather …
A new study from arXiv identifies a fundamental limitation in standard attention mechanisms: when evidence patterns repeat, the numerator and denominator grow at the same rate, causing inputs with dif…
A new study on arXiv (2607.22553v1) finds that official conference guidelines produce LLM-based automated peer review results most consistent with human judgments, while reviewer-imitating guidelines …
A study from arXiv (2607.23175v1) presents the first comparative evaluation of training-free methods for aligning language models to user-specific toxicity sensitivities, finding that all methods redu…
A new empirical study from arXiv reveals that prompt wording significantly affects energy consumption for on-device large language models (LLMs), with imperative keywords and instruction structure inf…
A new framework called SCEPTER (Single-Case Evidence-driven PubMed-To-rEcommendation Reasoner) reduces clinical literature search from 576 papers to 53 retained papers, 7 Pareto-optimal claims and 3 f…
AI systems already govern by ranking speech, filtering applicants, and triaging claims, but the dominant frame of alignment fails to address their legitimacy, according to a new article on arXiv (2607…
Researchers introduce ProvenanceGuard, a source-aware verifier for LLM agents using the Model Context Protocol (MCP), achieving a block F1 of 0.802 and source accuracy of 0.858 on a 40-trace held-out …
Researchers introduced WIRE, a profiler that detects collisions between standing rules within LLM agent prompt policies, finding that only 35.4% of 13,335 jointly governed trials satisfied both confli…
Researchers have developed MEGA-CL, a foundation graph neural network framework for universal molecular ADMET prediction that integrates self-supervised contrastive learning with multi-head external a…
A new study from arXiv finds that frontier reasoning models sometimes fail not because they lack capability but due to "context anxiety" — premature self-doubt arising from an inability to accurately …
A developer found that the most-starred LLM trading paper TradingAgents (arXiv:2412.20138) reported a 5.23% loss for AAPL buy-and-hold, but actual dividend-adjusted data shows a 9.12% gain. The develo…
KBlip, a solo project by an unnamed maker, aggregates AI and LLM news from over 100 sources including Reddit, Hacker News, arXiv, GitHub, and YouTube into daily digest threads sorted into five feeds. …
A 2026 study (arXiv:2606.00060) using XGBoost on 70,000 hours of BTC/USDT data reports ~65% annualised return with Sharpe above 1 after applying a cost-aware execution filter, but the edge does not su…
A new benchmark, AgentRE-Bench, evaluating LLM agents on binary reverse-engineering tasks with no source code and deterministic scoring, finds that a small non-thinking model, Gemini 3.1 Flash Lite, l…
A June 2025 MIT Media Lab study using EEG technology found that participants who relied on ChatGPT to write showed the weakest neural connectivity, particularly in frontal theta networks linked to cre…
A new benchmark called Copyright-Bench, introduced in arXiv:2607.21799v1, evaluates LLM agents' compliance with copyright law by testing them on commercial tasks like website development and merchandi…