Research

Tasks

Explore the problems researchers are working on and the benchmarks used to measure progress.

4,899 tasksPage 141 of 196

Task	Domain	Benchmarks	Papers	Last update
General Knowledge Task	llm chat instruction	1	1	Feb 22, 2026
Long-range Next-token prediction	llm chat instruction	1	1	May 6, 2026
Helpful Response Evaluation	llm chat instruction	1	2	Feb 19, 2026
Refusal Induction	adversarial safety securityllm chat instruction	1	1	Feb 22, 2026
Knowledge-Orthogonal Reasoning	llm chat instruction	1	1	May 6, 2026
Role Generalization	llm chat instruction	1	1	Feb 18, 2026
Failure localization	llm chat instruction	1	1	May 6, 2026
Fact recall	llm chat instruction	1	1	May 6, 2026
LLM Agent Defense Evaluation	adversarial safety securityagents tool use+1	1	1	May 6, 2026
Group Booking with failures (Grand Rollback)	llm chat instructionagents tool use	1	1	May 6, 2026
Response-Use classification	llm chat instructionnlp	1	1	Feb 22, 2026
Stealth Sycophancy Detection	llm chat instructionadversarial safety security	1	1	May 6, 2026
Zero-shot Language Modeling Evaluation	nlpllm chat instruction	1	1	Feb 22, 2026
Moral Steering	llm chat instruction	1	1	May 6, 2026
Fine-grained Moral Steering	llm chat instructionadversarial safety security	1	1	May 6, 2026
Multi-turn Medical Diagnosis	llm chat instructionbio medical	1	1	Feb 18, 2026
Multi-stage Reasoning and Navigation	robotics embodied aillm chat instruction	1	1	Feb 22, 2026
In-context comprehension	llm chat instruction	1	1	May 6, 2026
Engagement Evaluation	llm chat instruction	1	1	Feb 22, 2026
General Downstream Task Evaluation	llm chat instruction	1	1	May 6, 2026
Discriminative Accuracy	llm chat instruction	1	1	May 6, 2026
Perceived Authenticity Evaluation	llm chat instruction	1	1	Feb 22, 2026
Downstream Policy Evaluation	llm chat instruction	1	1	May 6, 2026
Overall Satisfaction Evaluation	llm chat instruction	1	1	Feb 22, 2026
Long-Paragraph Factuality	llm chat instructionnlp	1	1	Feb 18, 2026