AI Science Tracker
EN

The Hong Kong University of Science and Technology

2 discoveries from 2026

Discoveries

A benchmark finds retained experience can help agents but is often not updated after environment changes

Source check: Source checked. Not independent scientific confirmation.

Finding: Deterministic task graders measured performance and experience updating under simulated environment changes. These controlled benchmark results do not establish performance in open-ended or deployed environments.

Technology: The study compared 20 language-model agents with retained or cleared experience across task streams in 18 environments.

Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
The study compared 20 language-model agents with retained or cleared experience across task streams in 18 environments.
Test and result
Deterministic task graders measured performance and experience updating under simulated environment changes.
Limitation / still unknown
These controlled benchmark results do not establish performance in open-ended or deployed environments.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://arxiv.org/html/2611.05450v1

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Reasoning models usually inherited the revision marker “wait” before reinforcement learning

Source check: Source checked. Not independent scientific confirmation.

Finding: In 17 of 18 tested reinforcement-learning runs starting from models that rarely wrote “wait,” the word stayed rare even as math accuracy improved. In the traced OLMo-3 lineage, the word appeared during mid-training before reinforcement learning and became more frequent afterward.

Technology: OLMo-3, Qwen2.5, Qwen3, Mistral

Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
Researchers sampled language-model outputs across recorded training stages and externally released and newly trained reinforcement-learning checkpoints.
Test and result
The analysis compared published checkpoint histories and controlled training runs, using fewer than 0.05 occurrences per thousand generated tokens as its absence criterion.
Limitation / still unknown
The result concerns a visible marker on the sampled tasks. It does not show whether a model reconsidered an answer without writing “wait.”
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://arxiv.org/abs/2610.30882 · https://arxiv.org/html/2610.30882v1

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: linked separately.

Review date: 21 Nov 2026. Review scope: arXiv v1 affiliations, main provenance contrasts, released-checkpoint lineage and robustness appendix.

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open the interactive profile for the timeline, or browse all organizations and model families.