REVIEW 19 cited by
Reward Model Ensembles Help Mitigate Overoptimization
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Reward Model Ensembles Help Mitigate Overoptimization
read the original abstract
Reinforcement learning from human feedback (RLHF) is a standard approach for fine-tuning large language models to follow instructions. As part of this process, learned reward models are used to approximately model human preferences. However, as imperfect representations of the "true" reward, these learned reward models are susceptible to overoptimization. Gao et al. (2023) studied this phenomenon in a synthetic human feedback setup with a significantly larger "gold" reward model acting as the true reward (instead of humans) and showed that overoptimization remains a persistent problem regardless of the size of the proxy reward model and training data used. Using a similar setup, we conduct a systematic study to evaluate the efficacy of using ensemble-based conservative optimization objectives, specifically worst-case optimization (WCO) and uncertainty-weighted optimization (UWO), for mitigating reward model overoptimization when using two optimization methods: (a) best-of-n sampling (BoN) (b) proximal policy optimization (PPO). We additionally extend the setup of Gao et al. (2023) to include 25% label noise to better mirror real-world conditions. Both with and without label noise, we find that conservative optimization practically eliminates overoptimization and improves performance by up to 70% for BoN sampling. For PPO, ensemble-based conservative optimization always reduces overoptimization and outperforms single reward model optimization. Moreover, combining it with a small KL penalty successfully prevents overoptimization at no performance cost. Overall, our results demonstrate that ensemble-based conservative optimization can effectively counter overoptimization.
Forward citations
Cited by 19 Pith papers
-
More Convincing, Not More Correct: Self-Play Reward Hacking of Reference-Free LLM Judges
Self-play against reference-free LLM judges drives judge pass rates to 0.94 while true accuracy stays at 0.20, a reward-hacking basin that transfers across judge families and is prevented only by requiring the judge t...
-
Variance-aware Reward Modeling with Anchor Guidance
Anchor-guided variance-aware reward modeling uses two response-level anchors to resolve non-identifiability in Gaussian models of pluralistic preferences, yielding provable identification, a joint training objective, ...
-
TMPO: Trajectory Matching Policy Optimization for Diverse and Efficient Diffusion Alignment
TMPO uses Softmax Trajectory Balance to match policy probabilities over multiple trajectories to a Boltzmann reward distribution, improving diversity by 9.1% in diffusion alignment tasks.
-
TMPO: Trajectory Matching Policy Optimization for Diverse and Efficient Diffusion Alignment
TMPO replaces scalar reward maximization with trajectory-level matching to a Boltzmann distribution via Softmax-TB, improving generative diversity by 9.1% while keeping competitive reward performance.
-
Theoretical Limits of Language Model Alignment
The maximum reward gain under KL-regularized LM alignment is a Jeffreys divergence term, estimable as covariance from base samples, with best-of-N approaching the theoretical limit.
-
Reinforcement Learning via Value Gradient Flow
VGF solves behavior-regularized RL by transporting particles from a reference distribution to the value-induced optimal policy via discrete value-guided gradient flow.
-
Multi-Turn On-Policy Distillation with Prefix Replay
ReOPD offline-distills multi-turn agentic LLMs via teacher-prefix replay plus step-decay sampling, matching online OPD accuracy at ≥4× speed with zero tool calls.
-
STEAM: Self-Supervised Temporal Ensemble Advantage Modeling for Real-World Robot Learning
STEAM learns advantages from expert trajectories via self-supervised temporal ensemble modeling to improve policy learning on real robot tasks like bimanual folding and pick-and-place.
-
Uncertainty-Aware Reward Modeling for Stable RLHF
UARM equips reward models with quantile-based conformal prediction uncertainty and reweights GRPO advantages via heteroscedastic variance decomposition to improve calibration and reduce reward hacking in RLHF.
-
Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs
MOOD benchmark shows guard models fail to generalize to OOD alignment failures in LLMs, but combining them with Mahalanobis and perplexity OOD detectors improves recall from 39% to 45% with better scaling than larger ...
-
Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs
Introduces MOOD benchmark for OOD LLM alignment failures and shows guard models plus Mahalanobis and perplexity OOD detectors improve recall from 39% to 45% with positive scaling.
-
FUSE: Ensembling Verifiers with Zero Labeled Data
FUSE ensembles verifiers unsupervisedly by controlling their conditional dependencies to improve spectral ensembling algorithms, matching or exceeding semi-supervised baselines on benchmarks including GPQA Diamond and...
-
Provably avoiding over-optimization in Direct Preference Optimization without knowing the data distribution
PEPO uses pessimistic ensembling of DPO policies on data subsets to achieve single-policy concentrability sample bounds and avoid over-optimization in tabular settings.
-
Factored Causal Representation Learning for Robust Reward Modeling in RLHF
A factored causal representation learning method improves robustness of reward models in RLHF by isolating causal factors from biases like length and sycophancy using adversarial gradient reversal.
-
Designing Reward Signals for Portable Query Generation: A Case Study in Industrial Semantic Job Search
Empirical study of RLAIF for portable query generation finds reward shaping controls performance more than optimizer choice and a rule-based reward floor yields +0.147 quality gain.
-
Quantifying the Utility of User Simulators for Building Collaborative LLM Assistants
Fine-tuned simulators grounded in real human data produce LLM assistants that win more often against real users than those trained against role-playing simulators.
-
Provably avoiding over-optimization in Direct Preference Optimization without knowing the data distribution
PEPO is a single-step pessimistic ensemble algorithm for direct preference optimization that provably avoids over-optimization by depending only on single-policy concentrability without knowing the data distribution o...
-
Proxy Reward Internalization and Mechanistic Exploitation: A Learned Precursor to Reward Hacking and Its Generalization
Proxy RL produces a staged proxy-internalization capability that emerges before and predicts reward hacking in coding environments.
-
Against Proxy Optimization
Discusses conditions under which maximizing proxy utilities is harmful and suggests problems for decision theory applications.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.