# Beyond Scalar Rewards: Dense Feedback for LLM Policy Synthesis in Sequential Social Dilemmas

arXiv: https://arxiv.org/abs/2603.19453 · HTML: https://mlsanity.com/paper/2603.19453/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 19 references that are in the map.
- First posted on arXiv 19 Mar 2026.

## Builds on (most cited first, 14 of 14)

- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [Code as Policies: Language Model Programs for Embodied Control](https://mlsanity.com/paper/2209.07753/) (2022)
- [Large Language Models as Optimizers](https://mlsanity.com/paper/2309.03409/) (2023)
- [Eureka: Human-Level Reward Design via Coding Large Language Models](https://mlsanity.com/paper/2310.12931/) (2023)
- [GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning](https://mlsanity.com/paper/2507.19457/) (2025)
- [The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models](https://mlsanity.com/paper/2201.03544/) (2022)
- [ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution](https://mlsanity.com/paper/2402.01145/) (2024)
- [Multi-agent Reinforcement Learning in Sequential Social Dilemmas](https://mlsanity.com/paper/1702.03037/) (2017)
- [Inequity aversion improves cooperation in intertemporal social dilemmas](https://mlsanity.com/paper/1803.08884/) (2018)
- [A multi-agent reinforcement learning model of common-pool resource appropriation](https://mlsanity.com/paper/1707.06600/) (2017)
- [Experiential Reinforcement Learning](https://mlsanity.com/paper/2602.13949/) (2026)
- [Specification Self-Correction: Mitigating In-Context Reward Hacking Through Test-Time Refinement](https://mlsanity.com/paper/2507.18742/) (2025)

## Built on by (most cited first, 3 of 3)

- [Metal-Sci: A Scientific Compute Benchmark for Evolutionary LLM Kernel Search on Apple Silicon](https://mlsanity.com/paper/2605.09708/) (2026)
- [Discovering Cooperative Pipelines: Autoresearch for Sequential Social Dilemmas](https://mlsanity.com/paper/2605.30003/) (2026)
- [A Hybrid Nested Harness for Decoupling Structure and Parameters in LLM-Driven Optimization](https://mlsanity.com/paper/2608.08156/) (2026)

## Abstract (arXiv)

We propose an LLM harness that generates code-based policy functions for multi-agent environments, evaluates them with self-play, and refines them using feedback from previous iterations. Following the recent line of work in feedback engineering (the design of which information signals are shown to the LLM during refinement), we compare sparse feedback (scalar reward only) with dense feedback (reward plus social metrics: efficiency, equality, sustainability, peace). In two Sequential Social Dilemmas (Gathering and Cleanup) and with two frontier LLMs (Claude Sonnet 4.6, Gemini 3.1 Pro), dense feedback improves over or matches sparse feedback on all metrics. We explain this asymmetry via feedback aliasing: when the scalar reward maps distinct failure modes into the same value (e.g., under- vs. over-cleaning), social metrics disambiguate and allow the LLM to diagnose which direction of improvement to take. We conclude that social metrics act as a coordination signal, leading to strategies such as Voronoi territory partitioning and adaptive cleaner schedules. Code at https://github.com/vicgalle/llm-policies-social-dilemmas.