Category
9,568 papers
The minimal viable subspace (MVS) method, which searches for the lowest-rank activation subspace at a layer that preserves task performance when everything outside that subspace is ablated, suggests a functionally localizable core machinery for logic akin to that in the human brain.
A four-stage LLM pipeline gives up as much as 40.5 accuracy points at its own interfaces (gemma-3-12B on MATH-500, Holm-corrected p = 1.66e-19; the largest tax in the primary family). We hold model, problem, stages, stage prompts and completion budget fixed, vary only whether each stage can still see the original probl...
It is shown that strong small-option performance does not necessarily imply robust large-scale candidate comparison and that hierarchical partitioning and permutation-based inference improve accuracy by roughly 20 percentage points at $N=160$ on both HotpotQA and MIMIC.
This work splits the reader's loss into a budget loss, which any state of that size must incur, and a write-time regret, which comes from the writer's choices, and teaches the writer to prefer the better ones.
Reach audiences
Advertise in front of researchers, engineers, and readers.
One variable sets what an upstream fault costs a staged pipeline of language-model agents: re-derivability, how much of what a stage needs it can rebuild from the original problem. Grounding an inspector agent in that problem is worth +0.608 [+0.517, +0.700] to +0.358 over a blind one on four open-weight backbones serv...
CLAIRE (Clinical Language and Agentic Intelligence for Reasoning and Entry), a hybrid workflow that separates field-state discovery, source-to-field mapping, deterministic validation, bounded correction, escalation, and audit tracing, is developed and evaluated.
The Adaptive Consistency Graph (ACG) is introduced, which incrementally organizes execution evidence and its provenance in a persistent graph, then constructs a temporary requirement-centered view for each decision under a bounded context budget.
Auditing publicly released descendants of distilled models spanning diverse post-training objectives, SCOUT consistently identifies the distillation source and tracing teacher-associated *syntactic signatures* along training trajectories reveals that they emerge during distillation and persist through subsequent prefer...
MALA, a fused attention primitive that preserves score access to every legal causal interaction and uses normalized contribution to allocate post-score computation, indicates that allocating post-score computation according to normalized attention contributions can retain the evaluated capabilities of FullAttn while re...
It is found that model pairs can learn to communicate the secret using only one bit of feedback indicating whether the receiver inferred it correctly, during inference with fixed parameters and no supplied codebook or encoding examples.
Information-Gain-Gated Self-Distillation (IGSD) is proposed, which verifies on-policy token proposals with environment feedback before distillation, and supports environment-verified hindsight as an effective approach to reliable action-level supervision for search agents.
CUE-Mem provides a testbed for memory systems that selectively retain, retrieve, and use subtle multimodal evidence, and provides a testbed for memory systems that selectively retain, retrieve, and use subtle multimodal evidence.
Natural Language Processing
Our most capable vision-language-action model enhances robot dexterity, reliability, and safety
A Blog post by Ai2 on Hugging Face
A Blog post by IBM Research on Hugging Face
Flint-interactive gives AI agents a language for creating interactive visualizations. Click a chart, filter data, highlight related content, and explore beyond the view the author originally chose. The post Flint-interactive: A language for creating interactive data visualizations appeared first on Microsoft Research.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
