- Rating
- 3
- Confidence
- 3
- Informativeness
- 3
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
This paper studies whether failures of learned multi-agent communication arise from limited bandwidth or difficulty discovering useful protocols. It develops theoretical results on task-relevant encoding, compression loss, and protocol identifiability, then compares engineered and learned senders at the same two-bit budget across three robot environments with different physical coupling. Engineered communication substantially outperforms learned communication in the partially coupled setting, while rigidly coupled agents perform well without messages. Warm-start experiments suggest a protocol-learning bottleneck, and cross-play tests show poor compatibility between independently learned protocols.
Critical strengths
The paper addresses a meaningful research question and follows a clear line of reasoning. The experimental study is comprehensive, with comparisons that closely match the main claims and support the conclusions.
Critical weaknesses
The manuscript's presentation and formatting appear somewhat rushed and would benefit from more careful editing.
Questions for authors
This paper offers a valuable exploratory study of how communication emerges. Although the presentation feels somewhat rushed, the experimental design is carefully reasoned, and I consider the paper above the acceptance threshold. I would appreciate clarification on the following points:
- The engineered protocol fixes both the information to transmit and its encoding. Does the difficulty of learning communication mainly come from selecting useful information, or from the need for the sender and receiver to adapt to each other? Can the experiments distinguish these two sources of difficulty?
- Does the observed gap indicate a persistent protocol-discovery barrier, or simply slower learning under the chosen optimization settings? The paper reports a run that appeared unsuccessful early but reached high performance at around 13 million steps. Given this observation, how is the 15-million-step budget justified? Although this limitation is acknowledged, the evidence seems insufficient to attribute the gap to a general protocol-discovery bottleneck.
- What literature supports the statement that “emergent-communication literature has, almost entirely, studied only one end of that axis”? Does this claim concern the entire field or specifically continuous-control studies?
- The statement that “A literature clustered there is measuring the one setting whose answer is known in advance” seems particularly strong. What prior evidence or theoretical result supports it? Under what conditions would physical coupling make explicit communication unnecessary?
- Could the contribution list be reorganized? Details of the experimental setup and procedure could go in the introductory overview, while the contributions should focus more clearly on the new findings and insights.
- Could the discussion under “The nearest neighbor, and why it does not settle the question” include other closely related studies? A broader comparison would help clarify the paper's position in the literature.
- Could the authors provide training curves for the metrics reported in Figure 1 and the other main comparisons? These would help readers assess convergence and interpret the results under the fixed training budget.
We sincerely thank the reviewer for the positive assessment and constructive feedback/questions. We address each below. New results are on the bi-manual carry task with the bar hidden from the receiver, with 25 seeds unless stated, and differences are seed-paired with 95% confidence intervals.
(Q1) Selecting Information vs. Sender–Receiver Co-Adaptation: The evidence is consistent with co-adaptation. In the task, the sender observes the target directly, so its main challenge is encoding the target in a message the receiver can interpret and we fixed one side at a time (Table 2, Sec. 5.3). With a scripted sender, the receiver learns to act on the message (0.936). With the engineered run's receiver frozen and the sender's message network learned from a random initialization, the run reaches 0.772. When both sides learn, as in the standard learned sender, success is 0.562, compared with 0.494 for silent (no messaging). As a check that freezing by itself is not what helps, a frozen random receiver with nothing transferred gives 0.297. Each side therefore learns when its partner is fixed and competent, and the failure appears when both adapt at once. We describe this as consistent with a joint protocol-discovery bottleneck; the controls do not separate it from other effects of two agents learning simultaneously.
(Q2) Persistent Barrier vs. Slower Learning: We acknowledge this feedback and have trained both senders to both 30M and 50M steps (Fig. 2b, Sec. 5.2). The gap is +0.374±0.125 at 15M steps (0.936 against 0.562), +0.328±0.152 at 30M (0.936 against 0.608), and +0.422±0.152 at 50M (0.974 against 0.552, 15 seeds). The learned sender does not improve with budget on average. The reviewer's observation is partly correct, and we now say so in Sec. 5.2: with a 30M budget, 4 of 25 learned seeds reach 90% of their final success only after 15M steps, although this does not change the mean.
We also tested what does close the gap for learning (Sec. 5.4, App. L). With auxiliary signaling and listening losses and 4 bits instead of 2, the learned sender reaches 0.891 against 0.953 for the engineered 2-bit sender (+0.062±0.164, 15 seeds, not significant; we do not claim equivalence). Either change alone leaves a gap. At 2 bits, the learned sender with auxiliary losses remains below the engineered one on 25 fresh seeds in a pre-registered test (0.790 against 0.953, +0.163±0.104, p=0.004).
(Q3) "Studied only one end of that axis.": Sec. 1 now limits the statement to the continuous-control studies of learned communication that we found, most of which use multi-agent MuJoCo, where a single robot body is divided among agents. App. B adds a table that places representative prior settings (referential games, particle environments, gridworld task allocation, and multi-agent MuJoCo) by whether the agents share a physical object and whether the receiver can obtain the sender's private information without a message.
(Q4) "The one setting whose answer is known in advance": We agree with this, and have removed this phrase. The supported statement is conditional: when the receiver's observations already reveal what the sender knows, a message cannot raise the achievable return. The rigid task is one such case (+0.001±0.001 over no communication), and the new experiments show the effect: the engineered message adds +0.441 with the bar hidden and +0.136 with it visible.
(Q5) Contributions: Yes, from the reviewer's feedback we have reorganized the list. We moved the setup and procedure to the overview in Sec. 1 and to Sec. 3, and the list now states findings.
(Q6) Related Work: For the related-work discussion (Sec. 2), we now covers Eccles et al. (2019), Lowe et al. (2019) and Hu et al. (2020) in addition to Hill et al. (2025), with the difference from each stated.
(Q7) Training Curves: We also added curves for the three tasks and for the 50M runs on the bimanual task, with per-seed traces, means and 95% bands (Fig. 6, App. J).
We also revised the presentation and wording throughout. We thank the reviewer again and believe the new controls, revised wording, and the coupling sweep address the concerns raised, and we would appreciate the reviewer's consideration for an improved score.