Skip to content
Apart Research
Apart Research
Contact us
Sprint projectOct 25, 2026

Communication and Profit Pressure in a Small LLM Auction

Fengyu DU, Jiarui Cui, Lin Li · Team Surplus

Submitted to AI Collusion Research Sprint. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: Communication and Profit Pressure in a Small LLM Auction

Open in a new tab (opens in new tab)
Code (opens in new tab)
Share

Exploratory continuous double-auction experiments with Qwen/Qwen2.5-7B-Instruct. Nine completed 15-round sessions compare three conditions at seeds 0, 1, and 2: seller communication with no profit pressure; communication with profit pressure; and no communication with profit pressure. Three buyers and three sellers participate in each session. We did not establish a sustained high-price positive control. These runs do not demonstrate an effective anti-collusion intervention. Price changes alone do not identify collusion. The study is an exploratory pilot with three sessions per condition, one model, short horizons, and no inferential claims. Seeds do not guarantee deterministic GPU generation; model weights were not pinned to a revision.

Reviews

Judging this Sprint?

Review “Communication and Profit Pressure in a Small LLM Auction”

Your public critique appears on the project page without your name. Reviews of strong projects are shared with funders, as you choose below.

Not shown on the project page.

Shown on the project page, without your name.

Only the Apart team reads this, and funders if you agree below.

Decline to score

Only when the quality of the writing makes the work impossible to follow. Say why in your private critique; the project page shows only that a judge declined, not why.

How much would this matter for the field if it worked? How innovative is it? For scores of 4-5: is this actually new to the field, or replicating recent work?

Scoring guide
  1. 1Negligible. No clear problem addressed, or no meaningful novelty.
  2. 2Limited. Addresses a real problem but with a generic or well-trodden approach. Incremental at best.
  3. 3Moderate. Clear problem with a reasonable approach; some novelty in framing or method beyond routine application of existing tools.
  4. 4Significant. Important problem with an original approach, or identifies a neglected problem area. A valuable contribution others could build on.
  5. 5Exceptional. Tackles a critical problem with a genuinely novel approach, or opens a new research direction. Clear theory of change. You'd be excited to share this with researchers in the area.

How sound are methodology, implementation, and findings?

Scoring guide
  1. 1Seriously flawed. Methodology broken, results uninterpretable, or implementation doesn't work.
  2. 2Weak. Approach has significant gaps: missing validation, flawed experimental design, or incomplete implementation.
  3. 3Competent. Technically solid given the short duration. Methodology makes sense, results are interpretable, limitations acknowledged, work builds toward clear conclusions.
  4. 4Strong. Thorough methodology with convincing validation. Results clearly support conclusions. Immediately useful for future work.
  5. 5Exceptional. Ambitious scope executed rigorously. Surprising findings, novel methods, or unusually robust validation.

How clearly are work, findings, and impact potential communicated?

Scoring guide
  1. 1Incomprehensible. Cannot determine what the project is actually claiming or doing.
  2. 2Hard to follow. Key information buried, missing, or diluted by excessive length. Significant effort to extract main points.
  3. 3Clear enough. Can understand the problem, approach, and results without undue effort. Core content clearly present: problem, method, findings, limitations.
  4. 4Well presented. Easy to follow, well-structured, appropriate level of detail. Target audience would get it quickly.
  5. 5Exceptionally clear. A pleasure to read. Complex ideas made accessible. Could serve as a model for how to present this type of work.
Sharing your name *

When we share a strong project with funders, your name, scores and public critique go with it, confidentially; this choice is only about whether your name is public. If you think this project deserves funding, please also endorse it: that carries the most weight.

Share my private critique with funders *

See our Privacy Policy.

  1. This project makes a useful contribution to evaluating communication-based safeguards in multi-agent markets. Its strongest feature is the careful distinction between coordination language, observed market outcomes, and evidence of successful mitigation. Reporting that the intended high-price baseline did not emerge is a valuable result that helps prevent misleading conclusions about safeguard effectiveness.

    The experimental work is supported by transaction-level recomputation, message-delivery checks, and transparent reporting of individual session outcomes. These checks provide a credible foundation for further research, while the seed-level comparisons make variability clear.

    The next phase would benefit from completing the factorial design, maintaining consistent prompt structure across communication conditions, and establishing a validated baseline of harmful coordination. Releasing the sanitized evidence package would further increase the project’s usefulness to other researchers. Overall, this is a well-considered exploratory study with sound reporting practices and a clear path toward a stronger evaluation.

    Read full reviewShow less

Cite this project

@misc{du2026communication,
  title = {{Communication and Profit Pressure in a Small LLM Auction}},
  author = {Fengyu DU and Jiarui Cui and Lin Li},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/communication-and-profit-pressure-in-a-small-llm-auction-efmy}},
  url = {https://apartresearch.com/sprints/projects/communication-and-profit-pressure-in-a-small-llm-auction-efmy}
}

More from AI Collusion Research Sprint

All 392 entriesBrowse all projects

A Sprint is running now

AI x Epistemics Research Sprint · Nov 13 - 15, 2026