Rethinking Model Efficiency: Multi-Agent Inference with Large Models

• URL: http://arxiv.org/abs/2604.04929v1
• Date: Mon, 06 Apr 2026 17:59:35 GMT
• Title: Rethinking Model Efficiency: Multi-Agent Inference with Large Models
• Abstract summary: We conduct a comprehensive analysis of the latency across different components of vision-language models (VLMs) on simulated data.<n>Experiment shows that a large model with fewer output tokens can be more efficient than a small model with a long output sequence.<n>We propose a multi-agent inference framework that keeps large models with short responses but transfers the key reasoning tokens from the small model when necessary.
• Score: 23.878724608444145
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Most vision-language models (VLMs) apply a large language model (LLM) as the decoder, where the response tokens are generated sequentially through autoregression. Therefore, the number of output tokens can be the bottleneck of the end-to-end latency. However, different models may require vastly different numbers of output tokens to achieve comparable performance. In this work, we conduct a comprehensive analysis of the latency across different components of VLMs on simulated data. The experiment shows that a large model with fewer output tokens can be more efficient than a small model with a long output sequence. The empirical study on diverse real-world benchmarks confirms the observation that a large model can achieve better or comparable performance as a small model with significantly fewer output tokens. To leverage the efficiency of large models, we propose a multi-agent inference framework that keeps large models with short responses but transfers the key reasoning tokens from the small model when necessary. The comparison on benchmark tasks demonstrates that by reusing the reasoning tokens from small models, it can help approach the performance of a large model with its own reasoning, which confirms the effectiveness of our proposal.

Related papers

• Before Thinking, Learn to Decide: Proactive Routing for Efficient Visual Reasoning [87.18590662029432]
  We propose PRP, a Proactive Routing Paradigm that enables early decision-making by jointly evaluating the competence of both the draft and target models.<n>Our draft Rating Learning (DRL) equips the draft model with an internal confidence estimator, while Joint Rating Learning (JRL) predicts how well the target model can handle a given query.<n>These ratings enable fine-grained, instance-level textbfProactive Routing and substantially accelerate inference without compromising overall performance.
  arXiv  Detail & Related papers  (2026-06-29T12:30:24Z)
• Every Step Counts: Decoding Trajectories as Authorship Fingerprints of dLLMs [63.82840470917859]
  We show that the decoding mechanism of dLLMs can be used as a powerful tool for model attribution.<n>We propose a novel information extraction scheme called the Directed Decoding Map (DDM), which captures structural relationships between decoding steps and better reveals model-specific behaviors.
  arXiv  Detail & Related papers  (2025-10-02T06:25:10Z)
• ReTok: Replacing Tokenizer to Enhance Representation Efficiency in Large Language Model [9.1108256816605]
  We propose a method to improve model representation and processing efficiency by replacing the tokenizers of large language models (LLMs) Our method can maintain the performance of the model after replacing the tokenizer, while significantly improving the decoding speed for long texts.
  arXiv  Detail & Related papers  (2024-10-06T03:01:07Z)
• Matryoshka Multimodal Models [92.41824727506751]
  We propose M3: Matryoshka Multimodal Models, which learns to represent visual content as nested sets of visual tokens. We find that COCO-style benchmarks only need around 9 visual tokens to obtain accuracy similar to that of using all 576 tokens.
  arXiv  Detail & Related papers  (2024-05-27T17:59:56Z)
• The Larger the Better? Improved LLM Code-Generation via Budget Reallocation [32.0844209512788]
  It is a common belief that large language models (LLMs) are better than smaller-sized ones. This begs the question: what happens when both models operate under the same budget? We analyze code generation LLMs of various sizes and make comparisons such as running a 70B model once vs. generating five outputs from a 13B model.
  arXiv  Detail & Related papers  (2024-03-31T15:55:49Z)
• Multi-Candidate Speculative Decoding [82.05519287513444]
  Large language models have shown impressive capabilities across a variety of NLP tasks, yet their generating text autoregressively is time-consuming. One way to speed them up is speculative decoding, which generates candidate segments from a fast draft model that is then verified in parallel by the target model. This paper proposes sampling multiple candidates from a draft model and then organising them in batches for verification. We design algorithms for efficient multi-candidate verification while maintaining the distribution of the target model.
  arXiv  Detail & Related papers  (2024-01-12T17:15:23Z)
• Dissecting Multimodality in VideoQA Transformer Models by Impairing Modality Fusion [54.33764537135906]
  VideoQA Transformer models demonstrate competitive performance on standard benchmarks. Do these models capture the rich multimodal structures and dynamics from video and text jointly? Are they achieving high scores by exploiting biases and spurious features?
  arXiv  Detail & Related papers  (2023-06-15T06:45:46Z)
• Scalable Performance Analysis for Vision-Language Models [26.45624201546282]
  Joint vision-language models have shown great performance over a diverse set of tasks. Our paper introduces a more scalable solution that relies on already annotated benchmarks. We confirm previous findings that CLIP behaves like a bag of words model and performs better with nouns and verbs.
  arXiv  Detail & Related papers  (2023-05-30T06:40:08Z)
• When Ensembling Smaller Models is More Efficient than Single Large Models [52.38997176317532]
  We show that ensembles can outperform single models with both higher accuracy and requiring fewer total FLOPs to compute. This presents an interesting observation that output diversity in ensembling can often be more efficient than training larger models.
  arXiv  Detail & Related papers  (2020-05-01T18:56:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.