--- title: "StepFun AI — works of 2026: 3 papers, the most cited first" description: "3 works of StepFun AI from 2026 in our library, the most cited first. Among them: HAUNTATTACK: When Attack Follows Reasoning as a…" canonical: "https://research.elamur.ai/free/en/companies/stepfun-ai/year/2026/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/companies/stepfun-ai/year/2026/) [home](https://md.research.elamur.ai/free/en/) / [institutions](https://md.research.elamur.ai/free/en/orgs/) / [companies](https://md.research.elamur.ai/free/en/companies/) / [StepFun AI](https://md.research.elamur.ai/free/en/companies/stepfun-ai/) / 2026 # Works of StepFun AI from 2026 3 works · the most cited first · [← StepFun AI](https://md.research.elamur.ai/free/en/companies/stepfun-ai/) 1. **[HAUNTATTACK: When Attack Follows Reasoning as a Shadow](https://md.research.elamur.ai/free/en/papers/ma-2026-hauntattack-when-attack-follows-reasoning/) [Findings of the Association for Computational Linguistics](https://md.research.elamur.ai/free/en/venues/findings/)**Jingyuan Ma, Rui Li, Zheng Li, et al.*2026 · 3 citations* 2. **[PRIME: A Process-Outcome Alignment Benchmark for Verifiable Reasoning in Mathematics and Engineering](https://md.research.elamur.ai/free/en/papers/wang-2026-prime-process-outcome-alignment-benchmark/) [Annual Meeting of the Association for Computational Linguistics (ACL)](https://md.research.elamur.ai/free/en/venues/acl/)**Xiangfeng Wang, Hangyu Guo, Yanlin Lai, et al.*2026* 3. **[Revisiting Entropy Regularization: Adaptive Coefficient Unlocks Its Potential for LLM Reinforcement Learning](https://md.research.elamur.ai/free/en/papers/zhang-2026-revisiting-entropy-regularization-adaptive-coefficient/) [Findings of the Association for Computational Linguistics](https://md.research.elamur.ai/free/en/venues/findings/)**Xiaoyun Zhang, Xiaojian Yuan, Di Huang, et al.*2026* - [← back to](https://md.research.elamur.ai/free/en/companies/stepfun-ai/) — StepFun AI