The share of true binders from hallucination designs fell between the Adaptyv rounds
Design category share of binders of Adaptyv EGFR competition, both rounds
Section 4.2, verbatim: 'Interestingly, hallucinated designs, despite becoming more preferred in the second round, had a lower share of true binders (from 40% to 12%, Figure 9 C).' These are shares of the binder set by design category, not hit rates: the number of designs submitted or tested in each category is printed only in Figure 9, which was not read. The five curated categories are diversified binders, optimized binders, physics-based, de novo and hallucination.
The paper states that hallucinated designs, which became more popular in round 2, supplied a smaller share of the true binders: 40% in round 1 against 12% in round 2. The denominators behind these shares are in Figure 9C, which was not read.
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
6 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Claimed field design_category_share_of_binders Context-only references | Crowdsourced Protein Design: Lessons From the Adaptyv EGFR Binder Competition (Cotet et al., bioRxiv 2025) Section 4.2; Figure 9 caption Version: bioRxiv preprint version 1 | not individually reviewed No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: SHA-256 of one retrieval of a mutable web page. The hash describes that retrieval, not a stable artifact. |
| Source location Section 4.2; Figure 9 caption Context-only references | Crowdsourced Protein Design: Lessons From the Adaptyv EGFR Binder Competition (Cotet et al., bioRxiv 2025) Section 4.2; Figure 9 caption Version: bioRxiv preprint version 1 | not individually reviewed No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: SHA-256 of one retrieval of a mutable web page. The hash describes that retrieval, not a stable artifact. |
| Claimed value Section 4.2, verbatim: 'Interestingly, hallucinated designs, despite becoming more preferred in the second round, had a lower share of true binders (from 40% to 12%, Figure 9 C).' These are shares of the binder set by design category, not hit rates: the number of designs submitted or tested in each category is printed only in Figure 9, which was not read. The five curated categories are diversified binders, optimized binders, physics-based, de novo and hallucination. Context-only references | Crowdsourced Protein Design: Lessons From the Adaptyv EGFR Binder Competition (Cotet et al., bioRxiv 2025) Section 4.2; Figure 9 caption Version: bioRxiv preprint version 1 | not individually reviewed No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: SHA-256 of one retrieval of a mutable web page. The hash describes that retrieval, not a stable artifact. |
| Description The paper states that hallucinated designs, which became more popular in round 2, supplied a smaller share of the true binders: 40% in round 1 against 12% in round 2. The denominators behind these shares are in Figure 9C, which was not read. Context-only references | Crowdsourced Protein Design: Lessons From the Adaptyv EGFR Binder Competition (Cotet et al., bioRxiv 2025) Section 4.2; Figure 9 caption Version: bioRxiv preprint version 1 | not individually reviewed No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: SHA-256 of one retrieval of a mutable web page. The hash describes that retrieval, not a stable artifact. |
| Relationship: subject uc-binder-20261022-data-adaptyv-egfr-both-rounds Context-only references | Crowdsourced Protein Design: Lessons From the Adaptyv EGFR Binder Competition (Cotet et al., bioRxiv 2025) Section 4.2; Figure 9 caption Version: bioRxiv preprint version 1 | not individually reviewed No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: SHA-256 of one retrieval of a mutable web page. The hash describes that retrieval, not a stable artifact. |
| Name The share of true binders from hallucination designs fell between the Adaptyv rounds Context-only references | Crowdsourced Protein Design: Lessons From the Adaptyv EGFR Binder Competition (Cotet et al., bioRxiv 2025) Section 4.2; Figure 9 caption Version: bioRxiv preprint version 1 | not individually reviewed No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: SHA-256 of one retrieval of a mutable web page. The hash describes that retrieval, not a stable artifact. |
Sources and history
Release 2026-10-22-1b7cf027e4f6 · Record review: source checked
1 source record and release history
- Crowdsourced Protein Design: Lessons From the Adaptyv EGFR Binder Competition (Cotet et al., bioRxiv 2025) · Original source · bioRxiv preprint version 1
Technical metadata and extraction receipts
Stable ID: uc-binder-20261022-claim-adaptyv-hallucination-share-of-binders
- field
- design_category_share_of_binders
- review
- artifact sha256: 17039d69ecc37f45f6c402511b8f4497ce631003689fa95ad1ab221f8f787a4f; method: source-hash-verification; independent-cell-check; ai-assisted-source-review; method note: Re-downloaded (HTTP 200, 493,217 bytes, the same byte count as the pinned retrieval) but the SHA-256 differs, as expected for a mutable bioRxiv page whose hash describes one retrieval. Every extracted value and every quoted sentence was then located verbatim in the new retrieval, so the difference is in page furniture and not in the evidence. One section attribution and one quotation were corrected.; note: Reviewed in docs/reviews/use-cases/binder-2026-10-22.md. Automated AI review of the source transcription only; it is not human scientific review and nothing here was reproduced experimentally.; retrieval url: https://www.biorxiv.org/content/10.1101/2025.04.17.648362v1.full; reviewed at: 2026-10-22T15:00:00Z; reviewer: claude; reviewer note: Claude review agent, independent of the extraction agent; no human review claimed
- source locator
- Section 4.2; Figure 9 caption
- value
- Section 4.2, verbatim: 'Interestingly, hallucinated designs, despite becoming more preferred in the second round, had a lower share of true binders (from 40% to 12%, Figure 9 C).' These are shares of the binder set by design category, not hit rates: the number of designs submitted or tested in each category is printed only in Figure 9, which was not read. The five curated categories are diversified binders, optimized binders, physics-based, de novo and hallucination.