AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

Social Agents: Collective Intelligence Improves LLM Predictions

Aanisha Bhattacharyya, Abhilekh Borah, Yaman Kumar Singla, Rajiv Ratn Shah, Changyou Chen, Balaji Krishnamurthy

ICLR 2026 · Multi-agent systems and games › LLM agents · approach: LLM multi-agent decision making

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Social Agents, a multi-agent LLM framework that mimics collective human decision making to improve predictions, based only on the abstract's motivation.

Problem

LLM predictions may be improved by harnessing the benefits of collective intelligence (diverse, independent, distributed judgments) that have historically outperformed individual experts.

Approach

Uses a society of LLM agents with diverse perspectives and independent judgments whose aggregated outputs are expected to outperform individual predictions, in the spirit of wisdom-of-crowds averaging.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
aggregate persona-conditioned agents instead of a single no-persona expertGPT-4o · Tweet engagementaccuracy (%): 70.27 -> 86.90Improved
aggregate personas instead of single promptGPT-4o · Webpage likabilityPearson r: 0.28 -> 0.74Improved
aggregate personasGPT-4o · Ad CTR (Creative)MAPE: 72.45 -> 47.60Improved
aggregate personasQwen3 32B · Ad CTR (Real Estate)10-way accuracy (%): 26.10 -> 23.40Hurt
aggregate personasLLaMA 3.2 90B Vision · Webpage likabilityaccuracy (%): 57.73 -> 56.70Hurt
aggregate personasLLaMA 3.3 70B · ROAS (Creative)PE@20: 0.31 -> 0.24Hurt
aggregate personasGPT-4o · Video memorabilitySpearman rho: 0.33 -> 0.41Improved
aggregate personasQwen3 32B · Video memorabilitySpearman rho: 0.33 -> 0.33No clear effect
aggregate personasGPT-4o · Tweet generationBLEU-4: 23.77 -> 31.71Improved
aggregate personasGPT-4o · Behavioral attribute Persuasionaccuracy (%): 61.53 -> 61.81No clear effect
aggregate personas vs repeated no-persona calls or multi-LLM ensembleSocial Agents · Ad CTR, webpage likabilityMAPE / alignmentImproved
increase number of personas beyond 10-20Social Agents · Ad CTRMAPENo 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