Pith. sign in

REVIEW 12 cited by

Routing to the Expert: Efficient Reward-guided Ensemble of Large Language Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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

arxiv 2311.08692 v1 pith:RBYLWJVR submitted 2023-11-15 cs.CL cs.LG

Keming Lu , Hongyi Yuan , Runji Lin , Junyang Lin , Zheng Yuan , Chang Zhou , Jingren Zhou This is my paper · ORCID
classification cs.CLcs.LG
keywords llmsmodelrewardroutingzootercomputationensembleexpertise
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The complementary potential of Large Language Models (LLM) assumes off-the-shelf LLMs have heterogeneous expertise in a wide range of domains and tasks so that an ensemble of LLMs can achieve consistently better performance. Existing ensemble methods for LLMs mainly focus on reward model ranking of outputs, leading to significant computation overhead. To combat this issue, we revisit the complementary potential of LLMs and further elaborate it by mining latent expertise with off-the-shelf reward models. We propose Zooter, a reward-guided routing method distilling rewards on training queries to train a routing function, which can precisely distribute each query to the LLM with expertise about it. We also integrate a tag-based label enhancement to mitigate noise from uncertainty when using rewards as silver supervision. Zooter shows computation efficiency in inference as it introduces only a minor computation overhead of a routing function compared with reward model ranking methods. We evaluate Zooter on a comprehensive benchmark collection with 26 subsets on different domains and tasks. Zooter outperforms the best single model on average and ranks first on 44% of tasks, even surpassing multiple reward model ranking methods.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 12 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Decentralized Aggregation of LLM Predictions via Wagering Mechanisms

    cs.AI 2026-07 accept novelty 6.5 of 10

    A leave-one-out wagering payout makes LLM aggregation weights equal expected score advantage, yielding DSIC predictions, decentralized wager learning, and performance matching centralized routers.

  2. Optimal-Agent-Selection: State-Aware Routing Framework for Efficient Multi-Agent Collaboration

    cs.AI 2025-11 conditional novelty 6.0 of 10

    A state-aware contrastive router that selects the most relevant agent at each step improves multi-agent LLM accuracy by up to 23.8% while using a fraction of the tokens of fixed-pipeline baselines.

  3. Quality-of-Service Aware LLM Routing for Edge Computing with Multiple Experts

    cs.NI 2025-08 conditional novelty 6.0 of 10

    A DRL router using graph attention state abstraction and QoS-aware rewards improves average QoS by up to 35.78% over four baselines in simulated edge LLM routing.

  4. Tiny Reward Models

    cs.CL 2025-07 conditional novelty 6.0 of 10

    TinyRM shows that 400M-parameter bidirectional masked language models, tuned with FLAN-style prompting, DoRA, and layer freezing, outperform a 70B reward model on RewardBench reasoning and come close on safety.

  5. BEST-Route: Adaptive LLM Routing with Test-Time Optimal Compute

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A routing system that chooses both the model and the number of samples per query to meet a quality threshold, yielding up to 60% cost savings.

  6. RewardAnything: Generalizable Principle-Following Reward Models

    cs.CL 2025-06 conditional novelty 6.0 of 10

    RewardAnything follows natural-language reward principles at inference time and, with the new RABench benchmark, demonstrates that principle-conditioned listwise training beats fixed-preference reward models on held-o...

  7. IRT-Router: Effective and Interpretable Multi-LLM Routing via Item Response Theory

    cs.AI 2025-06 conditional novelty 6.0 of 10

    An IRT-based router that models each LLM's latent ability and each query's difficulty outperforms RouterBench on cost-performance reward across ID and OOD benchmarks.

  8. One for All: Update Parameterized Knowledge Across Multiple Models

    cs.CL 2025-06 conditional novelty 6.0 of 10

    One fine-tuned small model plus an ensemble step can update a fact across multiple large language models with a single edit, outperforming separate per-model editing.

  9. Towards Generalized Routing: Model and Agent Orchestration for Adaptive and Efficient Inference

    cs.MA 2025-09 reject novelty 5.0 of 10

    MoMA routes each query to a specialized agent or to the cost-optimal LLM, using judge-trained performance scores, a Pareto cost frontier, and TOPSIS selection.

  10. Enabling Flexible Multi-LLM Integration for Scalable Knowledge Aggregation

    cs.CL 2025-05 conditional novelty 5.0 of 10

    Adaptive selection and dynamic weighted fusion of source LLMs reduces knowledge interference and improves target model accuracy compared to FuseLLM.

  11. LightRouter: Towards Efficient LLM Collaboration with Minimal Overhead

    cs.AI 2025-05 conditional novelty 5.0 of 10

    LightRouter uses short preview outputs to filter a pool of LLMs down to two, then aggregates their full responses, beating ensemble baselines and matching costlier models.

  12. Towards Efficient Multi-LLM Inference: Characterization and Analysis of LLM Routing and Hierarchical Techniques

    cs.LG 2025-06 unverdicted novelty 4.0 of 10

    A survey of LLM routing and hierarchical inference techniques that proposes an unvalidated unified evaluation metric called the Inference Efficiency Score.

Pith tools