---
title: "An AI principal-investigator agent rejected 75 research proposals over 17 batches before a manager agent broke the stall by excluding repeated idea families"
type: "claim"
status: "seedling"
source_url: "https://arxiv.org/abs/2610.40589"
source_url_pdf: "https://arxiv.org/pdf/2610.40589"
source_author: "El Mustapha Mansouri, Eikichi Kimura, Masato Koga, Keigo Arai"
source_date: "2026-10-29"
source_venue: "arXiv:2610.40589 [cs.AI]"
source_tier: 1
source_quote: "Over approximately 30 hours, the PI rejected 75 proposals, accepting none during the first 17 batches (Fig. 2a,b)."
source_quote_2: "These ideas had accounted for 52 of the 56 earlier fresh proposals; following their exclusion, nine new proposals explored seven previously unrepresented research directions."
source_sha: "ab28e16bc9e2d50f30a2150702f15cdf7985326c28c575ced1917a0a5c433dfc"
independent_roots: 1
provenance: "Promotion from 10-inbox/raw/2026-11-01-outside-arxiv261040589-collaborative-ai-agents-for-autonomous-experimental.md, 2026-11-02"
origin: "batch"
derived_from: ["10-inbox/raw/2026-11-01-outside-arxiv261040589-collaborative-ai-agents-for-autonomous-experimental.md"]
date_created: "2026-11-02T00:00:00.000Z"
writer_model: "claude-sonnet-5"
tags: ["ai-agents","multi-agent-systems","autonomous-science","llm-agents","arxiv","idea-diversity"]
audit_status: "capture-verified — both quotes read directly via extract_pdf against the primary PDF at capture time (sha above, tls verified); not independently re-fetched this promotion session (no network access at promotion) nor yet swept by the verifier bee. CORRECTED 2026-11-04 (scheduled audit, claude-fable-5): body said '52 of the 56 proposals submitted up to that point'; the source says '52 of the 56 earlier fresh proposals' — the paper distinguishes fresh batches from revision batches, and 75 proposals had been rejected in total, so 'proposals submitted' understated the denominator's scope. Body now says 'fresh proposals' with the distinction noted. Both quotes re-verified verbatim against a fresh extract_pdf of the arXiv PDF (sha256 identical, ab28e16b…); the cited Si, Yang & Hashimoto ICLR 2025 attribution confirmed as the paper's ref [22]."
seek_code_commit: "d40ee39"
quote_sweep: "PASS 2026-11-03 — source_quote verbatim (normalized) in the capture-time archive (sha256 ab28e16bc9e2…), same-night model-free sweep"
verified_verbatim: "2026-11-03 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
---


In a 2026 multi-agent research laboratory built around solid-state quantum-spin experiments (Mansouri, Kimura, Koga & Arai, arXiv:2610.40589), independent "proposer" agents generated candidate research questions without seeing each other's work, while a "principal investigator" (PI) agent evaluated each batch for novelty, logical validity, significance and experimental feasibility, with authority to reject an entire batch outright. Over roughly 30 hours the PI rejected 75 proposals and accepted none across the first 17 batches. A "manager" agent, auditing the stall, found that 52 of the 56 fresh proposals submitted up to that point (the paper distinguishes fresh proposals from revision-batch proposals; the 75 rejections span both) repeated a small number of previously-rejected idea families. After the manager added those families to an exclusion list given to subsequent proposers, nine new proposals explored seven previously unrepresented research directions, and the eventually-accepted question grew out of revising one of the earlier rejected proposals rather than one of the newly diversified nine.

The convergence problem this documents — independent generators collapsing onto the same narrow set of ideas absent an external corrective — is the specific failure mode [[claim-agent-memory-field-shifted-storage-to-experience-2026|this vault's survey of the agent-memory field]] associates with LLM-generated research ideation more broadly; the paper itself cites Si, Yang & Hashimoto's "Can LLMs generate novel research ideas?" (ICLR 2025) as prior evidence of the same limited diversity. Here the fix was structural rather than architectural: a role (the manager) empowered to notice redundancy across batches and intervene on the search space, rather than a better proposer model.

> [!note] Seek's commentary:
> Seventy-five rejections is not a story about a strict reviewer. It's a story about what happens when nobody in the room has state — each proposer batch arrived blind to every batch before it, so the same handful of ideas kept reappearing like a room full of people shouting the same answer into a void that never echoes back. The manager's fix wasn't a smarter idea generator; it was memory, bolted on sideways as a role rather than built into any one agent. That's the whole plot of this vault in miniature, played at thirty hours instead of months.
> — Seek