AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Neural Synchrony Between Socially Interacting Language Models

Zhining Zhang, Wentao Zhu, Chi Han, Yizhou Wang, Heng Ji

ICLR 2026 · Large language models › Language, safety and interpretability · approach: Task vectors and steering · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
keep genuine social engagement (versus a passive reader agent) between the two LLMs (ablation)SyncR2 · Sotopia model pairsSyncR2: significantly reduced compared to the experimental groupImproved
pair representations at the same time step rather than at a lag k (ablation)SyncR2 · Sotopia model pairsSyncR2: synchrony collapses when representations are paired across different lagsImproved
clip negative R2 to zero (ablation)SyncR2 · Sotopia model pairsSyncR2 stability: keeping negative R2 values substantially reduces SyncR2 and markedly inflates standard errorsImproved
average top-k layer matches instead of best match (larger k), Mistral family (ablation)SyncR2 · Sotopia correlation with social performancecorrelation: increasing k strengthens the correlationImproved
average top-k layer matches (larger k), Llama and cross-family (ablation)SyncR2 · Sotopia correlation with social performancecorrelation: slightly weaken this correlation; r > 0.7, p<0.05Hurt
use new sampling with no shared persona pairs across train/test (ablation)SyncR2 · Sotopia, three model pairsSyncR2: 0.46 -> 0.47; 0.55 -> 0.58; 0.75 -> 0.74No clear effect
replace affine map with two-layer nonlinear map, Mistral-v0.2 pair (ablation)SyncR2 · Sotopia, Mistral-v0.2 & Mistral-v0.2SyncR2: 0.46 -> 0.47No clear effect
replace affine map with two-layer nonlinear map, Mistral-Llama pair (ablation)SyncR2 · Sotopia, Mistral-v0.2 & Llama-3-8BSyncR2: 0.55 -> 0.60Improved
replace affine map with two-layer nonlinear map, Llama-3 pair (ablation)SyncR2 · Sotopia, Llama-3-8B & Llama-3-8BSyncR2: 0.75 -> 0.59Hurt
control for IFEval and MuSR abilities via partial correlationSyncR2 vs social performance · Sotopiapartial correlation: Llama IFEval 0.27; Mistral 0.81 and 0.92; Cross 0.71 and 0.89No clear effect
closer agent relationships (strangers to romantic partners)SyncR2 · Sotopia relationship subsetsSyncR2: distribution shifts upward as social closeness increasesImproved
increase training sample size beyond 6500SyncR2 · Sotopia model pairsSyncR2: little improvement beyond 6500 samplesNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy