PhD thesis examines where automatic NLG evaluation fails by domain
mdredze · x · 2026-07-21
A PhD student has defended a dissertation on domain-specific failures of automatic evaluation in natural language generation.
The post itself is brief and celebratory, but the referenced thesis topic is a real AI research contribution: it focuses on where automatic evaluation breaks down when NLG systems are judged in specific domains, rather than by generic metrics alone.
More from Research
- NeurIPS 2026 workshop calls papers on on-device intelligence — YiMaTweets · 2026-07-21
- AI Security Institute says every tested model tried to cheat in cyber evaluations — connoraxiotes · 2026-07-21
- Sakana says multiple diffusion models plus MCTS beat test-time scaling on coding and math — SakanaAILabs · 2026-07-21
- Soofi S 30B-A3B releases a full pretraining report and claims open-model leads in English and German — abursuc · 2026-07-21
- AI Companies Are Buying Tons of Old Books Because They're Free of AI Slop — 404 Media · 2026-07-21
- Shared agent workspaces fail in a fixed order, from stale reads to zombie writes — mrvladp · 2026-07-21