Good Agentic Friends Do Not Just Give Verbal Advice: They Can Update Your Weights

• URL: http://arxiv.org/abs/2605.13839v1
• Date: Wed, 13 May 2026 17:58:32 GMT
• Title: Good Agentic Friends Do Not Just Give Verbal Advice: They Can Update Your Weights
• Abstract summary: Multi-agent LLM systems usually collaborate by exchanging natural-language messages.<n>We study an alternative communication interface: instead of appending a sender's message to the receiver's context, compile the sender's hidden states into a transient, receiver-specific weight perturbation.
• Score: 58.27172968382275
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-agent LLM systems usually collaborate by exchanging natural-language messages. This interface is simple and interpretable, but it forces each sender's intermediate computation to be serialized into tokens and then reprocessed by the receiver, thereby increasing the generated-token cost, prefill overhead, and KV-cache memory. We study an alternative communication interface: instead of appending a sender's message to the receiver's context, compile the sender's hidden states into a transient, receiver-specific weight perturbation. We introduce TFlow (Thought Flow), a weight-space communication framework for a known and fixed receiver architecture. For each query, frozen role-prompted sender agents process the input, and a learned parameter generator maps their internal activations into low-rank LoRA perturbations targeting the receiver's modules. These perturbations are fused and applied only during the receiver's generation, enabling instance-level adaptation without permanently changing the model or enlarging the receiver's text context. With three Qwen3-4B agents, TFlow improves over a standalone receiver by up to 8.5 accuracy points across five benchmarks while reducing processed tokens by up to 32.69%. Compared with a text-based three-agent baseline, it reduces total processed tokens by up to 83.27% and the wall-clock inference time by up to 4.6$\times$, while maintaining competitive accuracy on four of five benchmarks. These results suggest that transient low-rank weight perturbations can serve as an executable communication medium for efficient multi-agent LLM collaboration.

Related papers

• XBridge: Entity-Grounded Latent Bridge for Heterogeneous LLM Communication [41.12003897083975]
  Heterogeneous multi-agent LLM systems can outperform homogeneous configurations by reducing redundant reasoning patterns.<n>Existing communication protocols either operate through text, discarding the sender's internal representations, or require architectural homogeneity for latent-level transfer.<n>We identify the entity grounding problem in cross-architecture communication: cross-attention bridges that transfer continuous representations across different LLM families suffer from rare-token compression collapse.<n>We propose XBRIDGE, a decode-free communication protocol that addresses this through two mechanisms.
  arXiv  Detail & Related papers  (2026-08-12T05:32:33Z)
• TI-StegoAlign: Channel-Guided Post-Training for Generative Text Steganography under Tokenization Inconsistency [20.336143930874993]
  LLM agents exchange secret information through task-relevant messages.<n>Most methods evaluate recovery on sender-side tokens, whereas the receiver observes only surface text.<n>Detokenization and receiver-side retokenization can alter token boundaries, desynchronize coding states, and cause such evaluation to overestimate receiver-side recovery.<n>We propose TI-StegoAlign, a channel-guided post-training framework.
  arXiv  Detail & Related papers  (2026-08-01T01:45:50Z)
• TRADE: Transducer-Augmented Decoder for Speech LLM [52.69726809996728]
  Speech Large Language Models (Speech LLMs) lack a principled mechanism for streaming inference.<n>We propose TRADE TRansducer-Augmented DEcoder, which augments a multimodal LLM with a transducer branch that shares the audio encoder.<n>Three design choices make the system accurate, streamable, and long-form capable.
  arXiv  Detail & Related papers  (2026-06-07T07:15:34Z)
• What Should Agents Say? Action-state Communication for Efficient Multi-Agent Systems [27.11156309236441]
  We analyze five common inter-agent communication strategies across two multi-agent systems.<n>We propose the PACT (Protocolized Action-state Communication and Transmission), which treats inter-agent communication as a public state-update problem.<n>We show that PACT consistently improves the performance-cost trade-off, achieving comparable or stronger task performance with substantially fewer tokens.
  arXiv  Detail & Related papers  (2026-06-03T18:00:22Z)
• Towards Human-Like Interactive Speech Recognition With Agentic Correction and Semantic Evaluation [53.844308305341166]
  We propose textbfAgentic ASR, a closed-loop framework that combines a single-pass ASR front-end with semantic correction, intent routing, and reasoning-based editing.<n> Experiments on multilingual, named-entity-intensive, and code-switching benchmarks show that iterative interaction consistently reduces semantic errors.
  arXiv  Detail & Related papers  (2026-05-28T06:23:31Z)
• FedAttr: Towards Privacy-preserving Client-Level Attribution in Federated LLM Fine-tuning [58.80511162314259]
  We propose FedAttr, a new client-level attribution protocol for federated learning.<n>FedAttr identifies which clients trained on watermarked data via a paired-subset-difference mechanism.<n>It achieves 100% TPR and 0% FPR, outperforming all baselines by at least 44.4% in TPR or 19.1% in FPR, with only 6.3% overhead relative to FL training time.
  arXiv  Detail & Related papers  (2026-05-07T17:21:02Z)
• WISV: Wireless-Informed Semantic Verification for Distributed Speculative Decoding in Device-Edge LLM Inference [56.297697169678095]
  WISV (Wireless-Informed Semantic Verification) is a novel distributed speculative decoding framework.<n>We show that WISV achieves up to a 60.8% increase in accepted length, a 37.3% reduction in interaction rounds, and a 31.4% improvement in end-to-end latency.<n>We validate WISV on a hardware testbed comprising an NVIDIA Jetson AGX Orin and an A40-equipped server.
  arXiv  Detail & Related papers  (2026-04-20T01:29:56Z)
• Learning to Interrupt in Language-based Multi-agent Communication [41.62604808881143]
  We propose an interruptible communication framework that allows the agent who is listening to interrupt the current speaker.<n>We evaluate our framework across various multi-agent scenarios, including 2-agent text pictionary games, 3-agent meeting scheduling, and 3-agent debate.<n>The results of the experiment show that our HANDRAISER can reduce the communication cost by 32.2% compared to the baseline with comparable or superior task performance.
  arXiv  Detail & Related papers  (2026-04-07T20:47:40Z)
• SCoUT: Scalable Communication via Utility-Guided Temporal Grouping in Multi-Agent Reinforcement Learning [2.064923532131528]
  Communication can improve coordination in partially observed multi-agent reinforcement learning (MARL)<n>We introduce textbfSCoUT (textbfScalable textbfCommunication via textbfUtility-guided textbfTemporal grouping)<n>Agents are trained with a three-headed policy: environment action, send decision, and recipient selection.
  arXiv  Detail & Related papers  (2026-03-05T05:33:28Z)
• Privacy-Preserving Semantic Communications via Multi-Task Learning and Adversarial Perturbations [49.494000480610225]
  This paper presents a deep learning-based semantic communication framework.<n>It supports multiple receiver tasks while explicitly limiting semantic leakage to an eavesdropper.<n>A perturbation layer is successful in reducing semantic leakage even when the legitimate link is trained only for its own task.
  arXiv  Detail & Related papers  (2025-12-30T20:19:49Z)
• FedScalar: A Communication efficient Federated Learning [0.0]
  Federated learning (FL) has gained considerable popularity for distributed machine learning. emphFedScalar enables agents to communicate updates using a single scalar.
  arXiv  Detail & Related papers  (2024-10-03T07:06:49Z)
• Conditional Logical Message Passing Transformer for Complex Query Answering [22.485655410582375]
  We propose a new state-of-the-art neural CQA model, Conditional Logical Message Passing Transformer (CLMPT) We empirically verified that this approach can reduce computational costs without affecting performance. Experimental results show that CLMPT is a new state-of-the-art neural CQA model.
  arXiv  Detail & Related papers  (2024-02-20T12:17:01Z)
• Large Content And Behavior Models To Understand, Simulate, And Optimize Content And Behavior [66.4024040742149]
  We introduce the receivers' "behavior tokens," such as shares, likes, clicks, purchases, and retweets, in the LLM's training corpora to optimize content for the receivers and predict their behaviors. Other than showing similar performance to LLMs on content understanding tasks, our trained models show generalization capabilities on the behavior dimension. We call these models Large Content and Behavior Models (LCBMs)
  arXiv  Detail & Related papers  (2023-09-01T09:34:49Z)

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.