[KongchangAI]
ConceptPersonalized Memory Rubrics / 个性化记忆评分标准

PMR

A personalized credit assignment framework for long-term memory AI agents that decomposes terminal feedback into token-level fine-grained signals through a two-tier scoring rubric (a general layer and a user-adaptive layer), enabling memory policy optimization with zero inference overhead.

Timeline (last 90 days)

Nov 11

PMR (Personalized Memory Rubric) is a framework that converts terminal feedback into fine-grained, step-by-step, per-user credit assignment.

Unverified50%
Nov 11

PMR decomposes evaluation into two layers: a general layer that captures failure modes across user types, and a per-user adaptive layer that models individual characteristics as specialized increments on top of the general baseline.

Unverified50%
Nov 11

Through rubric-guided credit assignment, PMR redistributes credit at the token level: selectively amplifying verified decisions on successful trajectories and falling back to standard policy optimization on failed trajectories.

Unverified50%
Nov 11

PMR's rubrics and memory policy co-evolve in a closed loop, with anti-cycling guarantees.

Unverified50%
Nov 11

At deployment, PMR requires no rubrics or external judges, achieving zero inference overhead.

Unverified50%
Nov 11

Extensive experiments across four benchmarks show that PMR achieves consistent performance improvements over strong memory-augmented baselines.

Unverified50%
Nov 11

PMR's rubrics are not manually set but co-evolve with the memory policy: the general layer captures common failure modes across users, while the user-adaptive layer learns individual preferences from a small amount of user interaction data.

Unverified50%

All Facts (7)

Source Articles