arXiv cs.AI
·
16 Oct 2026
EvolvingBrain is not really a new brain. It is a runtime compounding layer that converts trajectories into reusable machine capital: procedural patterns, failure patterns, hierarchical skills, and state-aware execution rules.
arXiv cs.AI
·
16 Oct 2026
This paper replaces the closed-world, iid view of AI with a task-sequence view. It identifies three requirements for open-world competence: inference-time learning, rich representations, and predictive disentanglement. Its empirical clai...
arXiv cs.AI
·
16 Oct 2026
ARIA is a control-plane system for autonomous cognitive labor. It converts failed execution traces into verified repair data, then trains a compact model to detect and correct failures during live runs. Its real function is to reduce the...
arXiv cs.AI
·
16 Oct 2026
This paper converts a real engineering failure mode—distribution shift caused by a faster-changing environment—into a proposed scaling axis. It defines effective Speed of Coherent Cognition, trains agents at baseline reality speed, then ...
arXiv cs.AI
·
16 Oct 2026
This paper builds a tighter cognitive production loop: the LLM generates formulas, a learned verifier scores structural similarity, numerical error guides selection, and an analyst supplies diagnostic revisions. The important result is n...
arXiv cs.AI
·
16 Oct 2026
The paper packages a narrow but consequential form of mathematical labor into an automated pipeline: an LLM proposes structure, symbolic engines generate exact certificates, Monte Carlo Tree Search selects lemma-level moves, and Lean ver...
arXiv cs.AI
·
16 Oct 2026
CNRF is a representation-level triage system for VLM safety. It separates two signals that are often conflated: harmfulness and evidentiary support. It then selectively reviews suspicious requests and constrains responses instead of reje...
arXiv cs.AI
·
16 Oct 2026
AlphaDiana strips the mystique from “agents” and treats them as measurable execution systems. Its real contribution is separating model capability from orchestration, tools, skills, memory, traces, correctness, and resource use. It is an...
arXiv cs.AI
·
16 Oct 2026
This paper is not measuring whether AI has become safe, governable, or economically containable. It is repairing a failing instrument. Once benchmark scores saturate, it mines transcripts of already-correct answers—especially reasoning a...
arXiv cs.AI
·
16 Oct 2026
The paper turns latency from an implementation nuisance into the central evaluation variable. RealtimeBench keeps six classes of games running during inference, exposing the gap between capability measured in paused environments and capa...
arXiv cs.AI
·
15 Oct 2026
This is not a breakthrough in intelligence. It is a control-plane architecture for assembling existing capabilities—planning, machine learning, constraint programming, and language models—into executable graphs.
arXiv cs.AI
·
15 Oct 2026
This paper is building a memory prosthesis for industrial plants. Its real product is not cleaner maintenance text; it is the conversion of technician experience into verified, queryable maintenance capital. Rule-first extraction, proven...
arXiv cs.AI
·
15 Oct 2026
GroundGUI is not merely improving GUI interaction. It is converting the visual interface—the last human-readable layer of software—into a machine-operable surface. ScreenRef supplies the labeled perception data; D2RL closes the loop betw...
arXiv cs.AI
·
15 Oct 2026
This is a narrow benchmark study of two tool-discovery orderings. Its real move is to treat agentic competence as an information-routing problem: retrieve tools from the user’s wording, or first generate abstract capability hypotheses an...
arXiv cs.AI
·
15 Oct 2026
RLPE is an amortized optimizer for cognitive control. It observes a task model’s failures, learns which prompt interventions improve performance, and emits a reusable prompt once. The important innovation is not better wording. It is the...
arXiv cs.AI
·
15 Oct 2026
The paper dismantles accuracy as a universal certificate for multi-agent theory-of-mind modules. It replaces module-level scoring with deployment-level risk accounting: error costs, pairwise interactions, and correlated blind spots. Its ...
arXiv cs.AI
·
15 Oct 2026
This paper turns hidden LoRA behavior into a forensic search problem. SGES uses subspace alignment to find promising prompts without rendering everything; PVI then compares LoRA and base-model outputs. It makes undisclosed model speciali...
arXiv cs.AI
·
15 Oct 2026
This paper shifts symbolic regression from equation fitting to an agentic research loop. SCOPE runs parallel conjecture trajectories in which agents formulate questions, design experiments, build equations, exchange evidence, revise hypo...
arXiv cs.AI
·
15 Oct 2026
This abstract is a capability announcement disguised as a research report. It decomposes scientific work into a pipeline: drafting, ideation, planning, execution, aggregation, verification, ablation, analysis, and reporting. Once researc...
arXiv cs.AI
·
15 Oct 2026
The paper attempts to demote static inconsistency counts from “risk” to “suspected defects” requiring behavioral validation. Its empirical apparatus—575 skills, 1,182 manually validated inconsistencies, 49 tasks, three models, and 441 pa...
