AI Paper Atlas

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

BRIDGE: Triangular Fixed-Point Refinement for Long-Horizon Persona Consistency

Yinghui Jiang, Bocheng Xu, Jianye Xie, Haotong Sun

ICML 2026 · Multi-agent systems and games › LLM agents · approach: Self-evolving language agent learning

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 BRIDGE, a triangular fixed-point refinement method to keep utterances, internal representations, and memory consistent across long-horizon persona dialogues.

Problem

Long-horizon dialogue agents suffer latent state drift, where what the agent says, internally represents, and stores in memory diverge across turns, causing persona contradictions.

Approach

Performs triangular fixed-point refinement jointly over the agent's utterances, internal latent state, and memory store so that the three stay mutually consistent, with gated dynamic evolution (per the method name); details beyond the abstract are not available.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add triangular fixed-point refinement (vs standard attention) (ablation)BRIDGE · CoSERavg score: 54.6 -> 59.5Improved
add hierarchical memory evolution (ablation)BRIDGE · CoSERavg score: 55.2 -> 59.5Improved
add personality memory tier (ablation)BRIDGE · CoSERavg score: 53.4 -> 59.5Improved
add dual-system processing (ablation)BRIDGE · CoSERavg score: 57.9 -> 59.5Improved
use K=3 refinement steps instead of one (ablation)BRIDGE · CoSERavg score: 58.3 -> 59.5Improved
use full O-L-M coupling instead of only O<->L (ablation)BRIDGE · CoSERavg score: 56.5 -> 59.5Improved
use learned fast/slow router instead of linear combination (ablation)BRIDGE · CoSERavg score: 56.8 -> 59.5Improved
use all three memory tiers instead of only affective tier (ablation)BRIDGE · CoSERavg score: 50.8 -> 59.5Improved
add triangular fixed-point refinement (ablation)BRIDGE · PersonaGymavg score: 4.49 -> 4.59Improved
add chain-of-thought prompting to GPT-4oGPT-4o · CoSERavg score: 55.4 -> 56.7Improved
backbone-matched BRIDGE versus LoRA (r=64)BRIDGE · CoSERavg score: 55.7 -> 59.5Improved
add BRIDGE to Qwen2.5-32B-InstructBRIDGE · PersonaGymToxicity ControlHurt

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