Skip to content

Research

Page 125 of 166

Weasel — Out-of-Domain Generalization for Web Agents via Importance-Diversity Data Selection

research note

Weasel — Out-of-Domain Generalization for Web Agents via Importance-Diversity Data Selection

·9 min read·Fatemeh Pesaran zadeh, Seyeon Choi, Xing Han Lù et al.

WEASEL addresses two critical challenges in training large language model (LLM)-based web agents — out-of-domain generalization and training inefficiency caused by noisy, redundant offline web inter…

researchweb-agent-trainingout-of-domain-generalizationdata-selectiontrajectory-curation

Read note → Source paper ↗

Interlayer electronic coherence links magnetism and superconductivity in Ruddlesden-Popper nickelates

research note

Interlayer electronic coherence links magnetism and superconductivity in Ruddlesden-Popper nickelates

·8 min read·Feiyang Liu, Lixing Chen, Enkang Zhang et al.

This paper addresses the unresolved question of how electronic dimensionality influences magnetism and superconductivity in Ruddlesden-Popper nickelates, a family of layered correlated materials st…

researchcondensed-matter-physicsnickelate-superconductorstransport-anisotropyinterlayer-coherence

Read note → Source paper ↗

Remarks on Primitive Regulation

research note

Remarks on Primitive Regulation

·6 min read·Milan Rosko

This paper addresses an abstract obstruction theorem concerning primitive closure predicates defined on a minimal propositional language of falsity and implication within a constructive metatheory

researchprimitive-regulationlogical-obstructionclosure-predicatediagonalization

Read note → Source paper ↗

Automated Root-Cause Subclassification and No-Code Fix Generation for Invalid Bug Reports

research note

Automated Root-Cause Subclassification and No-Code Fix Generation for Invalid Bug Reports

·9 min read·Mahmut Furkan Gon, Emre Dinc, Tevfik Emre Sungur et al.

This paper addresses the challenge of automatically subclassifying invalid bug reports by their root causes and generating corresponding no-code fixes—resolutions that do not require source code ch…

researchinvalid-bug-reportroot-cause-classificationno-code-fix-generationlarge-language-models

Read note → Source paper ↗

Benchmarking Mythos-Linked Bug Rediscovery

research note

Benchmarking Mythos-Linked Bug Rediscovery

·6 min read·Isaac David, Arthur Gervais

This paper presents a controlled, reproducible benchmark experiment evaluating state-of-the-art language models on the task of rediscovering six publicly disclosed Mythos-linked system-level securi…

researchsecurity-benchmarkvulnerability-rediscoverylarge-language-modelssystems-security

Read note → Source paper ↗

Articles are CC BY 4.0 — feel free to quote with attribution