David Williams-King
2 indexed papers
Recent (6 mo)
2With code
0Influential cites
0Benchmarked
0Publications per year
226
Top categories
AI×2NLP×1ML×1Crypto×1
Frequent co-authors
Research Timeline
2026
FragBench: Cross-Session Attacks Hidden in Benign-Looking Fragments
The paper introduces FragBench, a novel benchmark designed to detect malicious LLM attacks that are split across multiple, seemingly benign sessions, showing that cross-session graph modeling is necessary for effective defense.
"I've Seen How This Goes": Characterizing Diversity via Progressive Conditional Surprise
The paper introduces the Decan metric, a novel, information-theoretic approach for measuring creative diversity in AI outputs, which successfully detects diversity loss across different model fine-tuning stages.
Highlighted terms show continued research focus across papers
Papers
cs.CLcs.AIcs.LGRecentJun 1, 2026
"I've Seen How This Goes": Characterizing Diversity via Progressive Conditional Surprise
The paper introduces the Decan metric, a novel, information-theoretic approach for measuring creative diversity in AI outputs, which successfully detects diversity loss across different model fine-tun…
View →
cs.CRcs.AIRecentMay 10, 2026