Metraipolis › archive › dc_2bae8cfdd4b5
J-Comm: Jacobian-Guided Latent Communication between LLM Agents — arXiv:2610.32871
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.32871
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-26T19:47:23.618Z
- SHA-256 of the archived text
0dccac0e74d77dd2d77d857fdf3113109188d9c9e3b01cf4650fa1defe16d0ae- Stable URL of this page
https://metraipolis.com/documento/dc_2bae8cfdd4b5
How to cite
J-Comm: Jacobian-Guided Latent Communication between LLM Agents — arXiv:2610.32871 — arXiv (en), https://arxiv.org/abs/2610.32871 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2bae8cfdd4b5 (Metraipolis archive dc_2bae8cfdd4b5, sha256 0dccac0e74d77dd2d77d857fdf3113109188d9c9e3b01cf4650fa1defe16d0ae)
BibTeX
@misc{dc2bae8cfdd4b5,
title = {J-Comm: Jacobian-Guided Latent Communication between LLM Agents — arXiv:2610.32871},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.32871},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2bae8cfdd4b5; sha256 0dccac0e74d77dd2d77d857fdf3113109188d9c9e3b01cf4650fa1defe16d0ae},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.32871. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Latent communication requires both an effective continuous reasoning process and an interface through which an external model can leverage its intermediate computation. We present J-Comm, a framework comprising two core components. First, a training-free Geometric Agreement for Latent Reasoning (GEAR) reweights token candidates over existing model embeddings, which monotonically optimizes geometric consensus among candidate tokens in embedding space at each step. Second, we introduce an H+J (Hidden-and-Jacobian) communication interface that gives the Receiver two complementary views of the Sender's reasoning. The H channel supplies continuous inputs derived from feedback distributions, while the J channel conveys intermediate-state cues in vocabulary-indexed Jacobian-lens coordinates. Intuitively, H provides inputs for the Receiver to reason from, while J offers additional guidance within its computation. The Receiver reconstructs these cues using its own dictionary and injects them into intermediate layers during prefill, enabling communication without sharing key-value caches or requiring identical model widths. In Qwen3-8B to Qwen3-4B communication, J-Comm achieves the highest mean score across seven benchmarks, including 94.80% on MATH500, 54.55% on GPQA-Diamond, 73.33% on AIME2024+2025, and 62.93% success on ALFWorld. Compared with TextMAS, it improves performance by 1.29-8.59 percentage points and reduces generated text tokens by 19.2-74.8%. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL) ; Multiagent Systems (cs.MA). Pagina del lavoro: https://arxiv.org/abs/2610.32871 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-26T19:45:18.257Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).