Fixed-profile soliton resolution fails (3D, 5D wave equation), with GPT-6 Astra
Awaiting review
The takeaway
Soliton resolution predicts that solutions of the focusing energy-critical wave equation split into finitely many fixed solitons plus radiation. Two preprints build bounded global solutions whose core keeps changing shape, contradicting the fixed-profile formulation of Duyckaerts, Kenig and Merle. The authors had the mechanism; GPT-6 Astra did the estimates.
Status
- Claim
Awaiting review
- Our reporting
- Medium confidence
- Verification
- Preprint only
- Importance
- 3 of 5
- Last verified
- 13 November 2026
Your AI and this story
- GPT-6 Astra196 days after its cutoff
- Claude Opus 5.5135 days after its cutoff
- Gemini 4 Hydrogen226 days after its cutoff
- Grok 4.7165 days after its cutoff
None of these four assistants can know about it. The closest, Claude Opus 5.5, stops 135 days before it.
Key facts
- Setting: focusing energy-critical wave equation u_tt − Δu = |u|^(4/(d−2)) u, nonradial, d = 3 (quintic) and d = 5; soliton resolution is known for radial data (Duyckaerts–Kenig–Merle; Jendrej–Lawrie in d ≥ 4) and along sequences for nonradial data (Duyckaerts–Jia–Kenig–Merle)
- Counterexample: radiation packets with growing gaps move the core along the conformal orbit of a sign-changing state, so it changes shape modulo translations, dilations and rotations; in 5D the limit set contains a continuum of classes
- What fails: ‘the fixed-profile soliton resolution conjecture’ — decompositions with a fixed finite collection of stationary profiles and fixed Lorentz velocities; resolution along sequences is not contradicted
- AI disclosure (5D, verbatim): ‘Inspired by the work discussed in Section 1.1, the authors asked whether radiation could drive a stationary core along the conformal orbit of a sign-changing state … A free unknown version of ChatGPT, assisted in developing the proof, and GPT-6 Astra worked out the estimates that justify this mechanism for the NLW.’
- Status: two arXiv preprints (math.AP), not refereed; no reaction found by 13 Nov
What happened
On 12 Nov 2026 Xuwen Chen, Justin Holmer and Shunlin Shen posted two papers constructing solutions of the energy-critical wave equation, in three and five space dimensions, whose solitary core never settles to one shape: weak, widely spaced bursts of radiation push it around a family of stationary states. This contradicts the version of the soliton resolution conjecture that fixes the soliton profiles in advance. The papers say the authors came up with the mechanism, a ChatGPT version helped develop the proof and GPT-6 Astra worked out the estimates.
Why it matters
Soliton resolution is a central programme in nonlinear dispersive equations, and the radial case was a celebrated theorem. The new results mark the limit of the strongest nonradial formulation. Both are unrefereed preprints.
What is disputed or not yet verified
| Verification | Preprint |
|---|
Sources
2 sources from 1 site. Numbers match the chips in the text.
2 sources: 2 primary
Primary
Changes
- Filed (science agent; both arXiv PDFs read)