docs: add external-research validation to PRIOR_ART.md - #4
Merged
Merged
Conversation
…entiator (PRIOR_ART.md)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Adds one new section to
docs/PRIOR_ART.md— "What the wider ecosystem confirms" — citing two external-research findings that validate existing project decisions. Docs-only change; no code touched.Additions
External validation of D27's "floor, not verified-true" framing. "Cited but Not Verified" (arXiv:2605.06635, May 2026) found frontier deep-research agents achieve only 39-77% factual/citation accuracy, and that accuracy drops as tool-call count scales up.
council/fidelity.py(D27 indocs/DECISIONS.md) already frames its citation-fidelity check as a lexical grounding floor — never a claim of verified truth. This paper is independent confirmation that framing is the honest, correct stance rather than an undersell.A stated differentiator: no comparable non-token peer network exists. Among local/decentralized-compute-sharing projects surveyed — GaiaNet (added staking/validator incentives in 2026), Petals (BitTorrent-style LLM sharding with no incentive layer, flagged in research literature as free-riding-vulnerable), and SwarmHarness (arXiv:2605.28764, a 2026 academic proposal for token/incentive-aligned agent networks) — none implement a non-transferable, no-token, earned-only credit model like D1 in
docs/DECISIONS.md. Stated explicitly as a genuine positioning differentiator.Verification
ruff check .— passed, no findings.pytest tests/ -q— 507 passed, 1 skipped (full suite green).🤖 Generated with Claude Code