arXiv AI
Pre-Registered Genesis Accounting for Recursive Self-Improvement: First Audited Operator-Genesis Events in a Physics-Discovery Runtime
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Oct 26
Minimum Description Length Unsupervisedly Separates Law-Grade from Statistical-Grade Knowledge in a Self-Improving Science Agent - and Reveals 86% of Its Own Discoveries Are Statistical
The paper reports on a self‑improving science agent that has admitted 252 scientific relations and uses Minimum Description Length (MDL) compression ratios as an unsupervised epistemic grader. The MDL‑ranked top discoveries match canonical physical laws (e.g., vis‑vis‑a, Balmer/Rydberg) with strong statistical enrichment and no counterexamples, while 86 % of the agent’s ledger entries compress poorly (cr≈1) and are identified as statistical regularities rather than true laws. The study introduces a dual‑axis bookkeeping system separating prediction value (R²) from compression value (cr), demonstrates its robustness, and presents falsifiable predictions.
arXiv AI
Jul 8
FirstResearch: Auditable Question Formation for LLM Scientific Discovery Agents
arXiv:2607. 05682v1 Announce Type: new Abstract: LLM systems for scientific discovery increasingly assist with ideation, literature synthesis, experiment planning, and report generation, but the first research question they propose can remain difficult to audit: it may sound plausible without exposing the mechanism, falsifier, or assumption that a scientist should inspect.
arXiv AI
Oct 7
Explore, Then Commit: Measurement-Efficient Scientific Law Discovery with Language Models
The paper presents an explore‑then‑commit protocol that uses a large language model to generate hypotheses, a programmatic planner to collect measurements, and a fresh prompt to synthesize a scientific law from fixed observations. In 576 NewtonBench trials across 12 physics modules, the protocol—especially when interpreter‑enabled planners are used—reduces the number of measurements needed and improves root‑mean‑squared logarithmic error for both GPT‑4.1‑mini and GPT‑4.1. The study demonstrates measurement savings in every module, though it notes that the causal components and generalization beyond noiseless direct‑equation tasks remain unresolved.
arXiv Machine Learning
Sep 17
Capability Emergence Can Be Forecast: Per-Seed, In Advance, With Calibrated Intervals, Certified False Alarms, and a Blind Pre-Registered Gate
The paper demonstrates that emergent capabilities in machine learning models can be forecasted with lead time, calibrated uncertainty, and controlled false‑alarm rates. Using per‑seed analysis on transformers, the authors show that the formation time of a previous‑token head predicts the emergence of an induction head with Spearman ρ = 0.977 and a median lead of 975 training steps. Conformal intervals, blind pre‑registered tests, and a multiplicative rule relating anchor and event times further validate the predictive framework across multiple model families and configurations.
arXiv AI
Jul 2
Testing Frontier Large Language Models' Physics Literacy in Parallel Physical Worlds
arXiv:2607. 00276v1 Announce Type: cross Abstract: Current large-language-model (LLM) physics benchmarks are usually scored by answer accuracy, which cannot distinguish genuine reasoning from recall of familiar problem patterns and reveals little about where a model's reasoning breaks down.
arXiv AI
Oct 27
Infinity Machine: a self-improving AI agent for research
The paper introduces the Infinity Machine, a self‑improving AI agent designed to solve research problems whose solutions can be programmatically verified. The system operates autonomously: it generates candidate solutions, verifies them, and audits the results without human intervention, all within a sandboxed environment governed by a kernel‑enforced gate. In a six‑day experiment, the Infinity Machine achieved four new state‑of‑the‑art results in quantum information and improved qubit‑routing and hardware efficiency, while also documenting several failure modes.