Unintentionally bad contexts (user suggestions or prior wrong assistant answers) cause LLMs to repeat errors, lose diversity, and flip stances, worsening with turns; RLVR on synthetic errors recovers 43–60% of the drop.
arXiv preprint arXiv:2402.15052 , year=
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 7roles
background 1polarities
background 1representative citing papers
The Social World Model supplies a five-dimension decomposition and closed-loop training loop that lets a 7B open model match Gemini 3 Flash on social metrics while showing zero forgetting on ASCENT-Bench.
DITTO uses RL with verbal feedback to train LLMs for human behavior simulation, reporting 36% average gains over base models and outperforming GPT-5.4 on 6 of 10 SOUL benchmark tasks.
A new framework identifies four mental model discrepancy types in team dialogues and demonstrates they carry predictive signals for future misalignments via uniform-weighted historical counts.
SiPeR improves recommendation accuracy and response quality in situated conversations by estimating scene transitions and performing Bayesian inverse inference with multimodal LLMs.
AIT Academy introduces a tripartite curriculum for AI agents across natural science, humanities, and social science domains, with reported gains of 15.9 points in security and 7 points in social reasoning under specific scheduling.
OSCToM uses RL-guided generation with an extended DSL and surrogate models to create nested belief conflict tasks, raising FANToM accuracy from 0.2% to 76% while being 6x more efficient.
citing papers explorer
-
Pigeonholing: how bad prompts hurt models, causing collapse and mistakes
Unintentionally bad contexts (user suggestions or prior wrong assistant answers) cause LLMs to repeat errors, lose diversity, and flip stances, worsening with turns; RLVR on synthetic errors recovers 43–60% of the drop.
-
Social World Model for Lifelong Social Intelligence
The Social World Model supplies a five-dimension decomposition and closed-loop training loop that lets a 7B open model match Gemini 3 Flash on social metrics while showing zero forgetting on ASCENT-Bench.
-
Reinforcing Human Behavior Simulation via Verbal Feedback
DITTO uses RL with verbal feedback to train LLMs for human behavior simulation, reporting 36% average gains over base models and outperforming GPT-5.4 on 6 of 10 SOUL benchmark tasks.
-
Are you with me? A Framework for Detecting Mental Model Discrepancies in Task-Based Team Dialogues
A new framework identifies four mental model discrepancy types in team dialogues and demonstrates they carry predictive signals for future misalignments via uniform-weighted historical counts.
-
Where and What: Reasoning Dynamic and Implicit Preferences in Situated Conversational Recommendation
SiPeR improves recommendation accuracy and response quality in situated conversations by estimating scene transitions and performing Bayesian inverse inference with multimodal LLMs.
-
AIT Academy: Cultivating the Complete Agent with a Confucian Three-Domain Curriculum
AIT Academy introduces a tripartite curriculum for AI agents across natural science, humanities, and social science domains, with reported gains of 15.9 points in security and 7 points in social reasoning under specific scheduling.
-
OSCToM: RL-Guided Adversarial Generation for High-Order Theory of Mind
OSCToM uses RL-guided generation with an extended DSL and surrogate models to create nested belief conflict tasks, raising FANToM accuracy from 0.2% to 76% while being 6x more efficient.