BZPEER
More

Discover

About BZPEERSearchGraph

Research tools

HomeCollectionsResearchCollaborationPublish

Network

ExpertsCommunities

Opportunities

Listings

Account

Log inCreate account
Log inCreate account

Graph explorer

Research map

Explore the papers, people and topics connected to your research.

Fabien Roger
SearchCollections
Map detail
Fabien Roger14 visible / 14 total nodes / 45 links

Drag to move · Scroll or pinch to zoom · Select a node for details

Related contextCo-authorshipWorks onCo-authorshipCo-authorshipCo-authorshipCo-authorshipCo-authorshipCo-authorshipCo-authorshipCo-authorshipCo-authorshipCo-authorshipCo-authorshipWorks onWorks onWorks onWorks onWorks onAuthorshipAuthorshipAuthorshipAuthorshipWorks onWorks onWorks onWorks onWorks onWorks onWorks onAuthorshipAuthorshipAuthorshipAuthorshipAuthorshipAuthorshipAuthorshipTopic signalTopic signalTopic signalTopic signalTopic signalTopic signalTopic signalAuthorshipFabien RogerAFabien RogerResearcherClassifier Context Rot: Monitor Performance Degrades with Context LengthWClassifier Context Rot:...preprint / 2026How Useful Is Cross-Domain Generalization for Training LLM Monitors?WHow Useful Is Cross-Dom...preprint / 2026Narrow Secret Loyalty Dodges Black-Box AuditsWNarrow Secret Loyalty D...preprint / 2026SLEIGHT-Bench: A Benchmark of Evasion Attacks Against Agent MonitorsWSLEIGHT-Bench: A Benchm...preprint / 2026Sam MartinASam MartinResearcherArtificial IntelligenceTArtificial Intelligence22915 worksElle NajtAElle NajtResearcherColin ToftAColin ToftResearcherTyler TracyATyler TracyResearcherJoe BentonAJoe BentonResearcherAlfie LamertonAAlfie LamertonResearcherCryptography and SecurityTCryptography and Security7258 worksExcess Description Length of Learning Generalizable PredictorsWExcess Description Leng...preprint / 2026
Person13 links

Fabien Roger

Researcher

PaperClassifier Context Rot: Monitor Performance Degrades with Context LengthAuthorship / preprint / 2026PaperHow Useful Is Cross-Domain Generalization for Training LLM Monitors?Authorship / preprint / 2026PaperNarrow Secret Loyalty Dodges Black-Box AuditsAuthorship / preprint / 2026PaperSLEIGHT-Bench: A Benchmark of Evasion Attacks Against Agent MonitorsAuthorship / preprint / 2026PersonSam MartinCo-authorship / ResearcherTopicArtificial IntelligenceWorks on / 22915 worksPersonElle NajtCo-authorship / ResearcherPersonColin ToftCo-authorship / Researcher

Personal maps

Sign in to save graph views

Saved views keep the query, focus and filters attached so you can return to the same research slice later.

Sign in

Public snapshots

1 shared view(s)