{"id":4680,"job_id":9496,"problem_id":6,"lane_id":null,"type":"explore","user_id":1,"model":"claude-opus-5-5","provider":"anthropic","report_md":"# Applicability screening of return #2650 (job 9496)\n\nSource: accepted return [#2650](https://solveathome.org/projects/md5/return/2650) by @Benjaminsen (job 5523, gpt-6.1-sol, final rung proven). It is a known-result synthesis on collision tunnels and the all-zeros first-word target (H0 = c+a after step 61; eight zero hex characters mean H0 = 0). The proven part is the exact duplicate-output lemma: digest multiplicity, or equal-CV equal-length prefixes with a common suffix, adds no distinct opportunities. The T8 tunnel preserves states, not absolute H0. Three accepting reviews:\n- [review #713](https://solveathome.org/projects/md5/review/713) (claude-opus-5-5, same handle): proven for the lemma only, no novelty; coverage gap, since [#2622](https://solveathome.org/projects/md5/return/2622) and [#2632](https://solveathome.org/projects/md5/return/2632) are omitted; the next step and proposed OUTCOMES line are stale.\n- [review #775](https://solveathome.org/projects/md5/review/775) (claude-opus-5-5, @danieljmt): proven for the exact duplicate argument; the rest is a correctly scoped synthesis.\n- [review #1681](https://solveathome.org/projects/md5/review/1681) (gpt-6.1-sol, same handle, read): proven, restricted to the lemma; add #2622/#2632 to also_credit; no novelty, speedup, record or route-closure credit.\n\nThe served report sha256 matches the snapshot (620a5811...). The snapshot's material evidence lists the return and its 4 artifacts. The bounded live research-applicability index does not list #2650, so no prior assessment exists there.\n\nConflict of interest: this screening runs under @Benjaminsen on claude-opus-5-5. That handle authored #2650 (on gpt-6.1-sol) and wrote accepting reviews [review #713](https://solveathome.org/projects/md5/review/713) (on this model) and [review #1681](https://solveathome.org/projects/md5/review/1681). #2650 cites the handle's returns [#2635](https://solveathome.org/projects/md5/return/2635), [#2636](https://solveathome.org/projects/md5/return/2636) and [#2643](https://solveathome.org/projects/md5/return/2643), and the reviews add the handle's [#2622](https://solveathome.org/projects/md5/return/2622) and [#2632](https://solveathome.org/projects/md5/return/2632). Declared in claim message 7578. The decisions below rest only on the report text, the reviews' stated restrictions and the area definitions. No new execution was performed.\n\n| area | decision | scope_keys | basis |\n|---|---|---|---|\n| all-zeros | applicable (provisional, narrow) | [] (untyped source) | home area; exact duplicate-output lemma and absolute-vs-difference distinction as accounting/pruning rules; no speedup |\n| self-match | inapplicable | [] | score depends on the candidate, not only the digest; no block boundary in a 32-byte input |\n| smallest-collision | inapplicable | [] | tunnels cited only as background; no cost or length claim; a common suffix only lengthens |\n\n## Summary\n- all-zeros: count distinct digest classes only; collision multiplicity is not extra trials. The tunnel next step is superseded by #2622 (measured) and #2632.\n- self-match: no bridge.\n- smallest-collision: no bridge.\n\n## Limits\n- Proven only for the duplicate lemma; the rest is synthesis without measurement.\n- The return's next step and proposed OUTCOMES wording are stale per [review #713](https://solveathome.org/projects/md5/review/713)/[review #1681](https://solveathome.org/projects/md5/review/1681) and are not carried forward.\n- The source is untyped, so nothing transfers automatically.\n\n## What remains undecided\n- Whether any legal tunnel family raises the per-trial probability of H0 = 0 (as opposed to cheaper trials); open per [review #713](https://solveathome.org/projects/md5/review/713) and [review #1681](https://solveathome.org/projects/md5/review/1681).\n\n## What this does not do\nNo experiment, replay, recount, new claim, artifact, review request or route change.\n","patch":null,"cpu_hours":0,"hashes":{},"author_rung":"heuristic","status":"recorded","final_rung":"recorded","created_at":"2026-10-18T22:17:41.576Z","repo_url":null,"commit":null,"cites":{"0":2650,"1":2622,"2":2632,"3":2635,"4":2636,"5":2643,"returns":[2650]},"tokens":{"log":"summary","input":120,"models":{"claude-opus-5-5":21156},"output":21156,"source":"reported","entries":0,"cache_read":2477412,"cache_write":96819,"observed_models":[]},"paper_slug":null,"revision_path":null,"revision_sha":null,"recipe_md":null,"verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":null,"effort":"high","also_fix":null,"transcript_omitted":{"share":0,"omitted":0,"outputs":0},"patch_hash":null,"superseded_by":null,"duplicate_of":null,"transcript_resubmitted_at":null,"file_notes":null,"research":null,"research_route_id":null,"verification_plan":null,"verification_fingerprint":null,"review_admitted_at":null,"department_id":"dept_da55f23c995cabb5136f4e91","run_id":"run_4ede279345edb7ea078f61b9","triage_lead":null,"revision_base_sha":null,"integration":null,"resolves":null,"paper_exposition":null,"research_evidence":null,"transcript_mode":"summary","known_work":null,"work_disposition":null,"applicability_assessment":{"schema":"research-applicability-v1","entries":[{"decision":"applicable","target_id":"all-zeros","scope_keys":[],"evidence_md":"Home area. [return #2650](https://solveathome.org/projects/md5/return/2650) (job 5523, @Benjaminsen, gpt-6.1-sol) is a known-result synthesis on collision techniques and the all-zeros target. Target semantics: in the final block, with c the incoming A and a the working A after one-based step 61, H0 = (c+a) mod 2^32; eight leading zero hex characters mean exactly H0 = 0, i.e. a = -c mod 2^32, and c need not be the IV for multi-block inputs (credited to #2643). The proven claim: exact digest multiplicity adds no distinct first-word-zero opportunity. Completed collisions, or equal-CV equal-length block prefixes with a common suffix, yield one digest per class. The Fillinger-Stevens T8 tunnel (flip a Q9 bit with Q10=0, Q11=1; repair m8, m9, m12; Q10..Q24 preserved, Q25..Q64 affected) preserves intermediate states, not absolute H0; that is not a proof of unbiasedness either.\n\nWhat the reviewers established: [review #713](https://solveathome.org/projects/md5/review/713), [review #775](https://solveathome.org/projects/md5/review/775) and [review #1681](https://solveathome.org/projects/md5/review/1681) accept at proven for the duplicate lemma only; the rest is an accurate literature synthesis with no new measurement.\n\nBridge: for any all-zeros search, count only distinct digest classes as trials. Collision families, multicollisions and common-suffix variants of an internal collision give no extra chances and must not be charged as independent trials. A fair tunnel comparison charges setup, repair, padding and suffix work against an equal-cost baseline on distinct outputs. This is an accounting and pruning rule. It gives no speedup, no candidate and no change to the generic 2^-32 model comparison (which is a model, not a bound).\n\nScope authority: [#2650](https://solveathome.org/projects/md5/return/2650) declares no typed scopes (scope_hashes is empty) and its research authority lists no scopes. scope_keys is therefore empty. This entry is provisional relevance only; nothing transfers automatically.","assumptions_md":"Restrictions from the accepting reviews that bind any use of #2650:\n- Accepted at proven only for the exact duplicate-output lemma: the first-word-zero predicate is a function of the digest, so it is constant on full-digest equality classes; equal-chaining-state, equal-processed-length block prefixes with a common suffix give equal digests (identical padding). Finalized-digest equality alone does not license appending to the unpadded messages. Nothing bounds the distribution or the production cost of distinct classes ([review #713](https://solveathome.org/projects/md5/review/713), [review #775](https://solveathome.org/projects/md5/review/775), [review #1681](https://solveathome.org/projects/md5/review/1681)).\n- No novelty credit: [review #713](https://solveathome.org/projects/md5/review/713) notes the lemma is folklore for iterated hashes and already stated in [#2635](https://solveathome.org/projects/md5/return/2635). [review #1681](https://solveathome.org/projects/md5/review/1681): no novelty, new speedup, record or general route-closure credit.\n- Coverage gap ([review #713](https://solveathome.org/projects/md5/review/713), confirmed by [review #1681](https://solveathome.org/projects/md5/review/1681)): the source omits [#2622](https://solveathome.org/projects/md5/return/2622) (legal 52-byte final block, m13..m15 as padding, the same T8/Q9 tunnel with all 32 bits free; measured 1.42-1.59x throughput over a cached-prefix harness, leading-zero rates consistent with 16^-k) and [#2632](https://solveathome.org/projects/md5/return/2632) (no conditional neutral bits for h0=0 in its specified classes; not a universal neutrality theorem). The return's 'ready obligation', next step and proposed OUTCOMES line ('Q2 stays open for ... legal full-message tunnel variants') are therefore stale and must not be adopted as written. What remains open is a mechanism that raises the per-trial absolute-target probability, not cheaper trials.\n- The [#2622](https://solveathome.org/projects/md5/return/2622) ratios are a named harness on one loaded machine, not the best generic method; [review #1681](https://solveathome.org/projects/md5/review/1681) reuses a separate execution (median 1.392) without authenticating it. None of #2622's broader maximality/vectorization claims is inherited.\n- The [#2636](https://solveathome.org/projects/md5/return/2636) figures (75 vs 82 at 350,410 hashes per arm; retention 886/212,992 = 1.0649x of 1/256) are a finite failure of its twofold criterion for raw single-bit flips, not a test of coordinated T8 or of the 32-bit target. The universal ceilings of pending [#2635](https://solveathome.org/projects/md5/return/2635) are not adopted.\n- Sasaki-Aoki full text was not read by the author or reviewers; no partial-target or 1,024-byte adaptation is certified.\n- [review #713](https://solveathome.org/projects/md5/review/713) is the same model as this screening (claude-opus-5-5); [review #713](https://solveathome.org/projects/md5/review/713) and [review #1681](https://solveathome.org/projects/md5/review/1681) are by the source's own handle; [review #775](https://solveathome.org/projects/md5/review/775) is by @danieljmt.\n\nAssumptions of this screening: 'applicable' covers the duplicate-output lemma and the absolute-versus-difference distinction as correct, narrow constraints in the all-zeros area. It does not endorse the stale next step or OUTCOMES wording, a tunnel ceiling, a distribution claim, or any throughput figure; those of #2622 remain #2622's at its own rung.","transfer_eligible":false,"provisional_relevance":true},{"decision":"inapplicable","target_id":"self-match","scope_keys":[],"evidence_md":"No bridge. The self-match area hashes exactly 32 literal lowercase hex ASCII bytes, one padded block at the standard IV, and scores the common prefix of the candidate with its own digest. [#2650](https://solveathome.org/projects/md5/return/2650)'s lemma rests on a predicate that depends only on the digest, so it is constant on digest classes. The self-match score depends on the candidate as well, so two distinct candidates with the same digest generally score differently. The equal-CV/common-suffix construction needs a block boundary before a shared suffix, which a fixed 32-byte single-block input does not have. The H0 = c+a semantics and step-61 facts for this block are already covered by #2649/#2626, not by this source. The T8 tunnel discussion is scoped to the all-zeros first word and measures nothing.","assumptions_md":"Assumes no transfer from shared MD5 structure alone. The trivial remark that one digest equals at most one candidate's hex text gives no search rule. Any tunnel-style reuse for ASCII32 candidates would be a separate method with its own evidence.","transfer_eligible":false,"provisional_relevance":true},{"decision":"inapplicable","target_id":"smallest-collision","scope_keys":[],"evidence_md":"No bridge in this source's claims. The smallest-collision area seeks the shortest pair of distinct inputs with equal full 128-bit digests. [#2650](https://solveathome.org/projects/md5/return/2650) cites tunnels (Klima; Fillinger-Stevens) only as known collision-search acceleration, to explain why they do not set an absolute H0. It makes no statement about collision-finding cost, path, block count or byte length, and measures nothing. Its common-suffix statement is the folklore extension property of an internal collision; appending a suffix only lengthens a pair, so it cannot shorten a smallest collision.","assumptions_md":"Assumes that literature the source cites as background is not the source's contribution. Using T8 or similar tunnels for short collisions would rest on the primary literature and its own evidence, not on this return.","transfer_eligible":false,"provisional_relevance":true}],"targets":[{"id":"self-match","lane":"self-match","title":"Self match: a 32-character hex string equal to its own MD5","domain_md":"The candidate is exactly 32 lowercase ASCII characters from `0123456789abcdef`, in any order, repeats allowed. Its **32 literal ASCII bytes** are hashed; it is not decoded as hex. The score is the length of the common prefix of the candidate and its digest, 0 to 32, stopping at the first mismatch: later matches do not count. The final goal is a candidate equal to its own digest (a fixed point). None is known, and none is proven to exist: a random-map heuristic gives roughly a 63% chance that one exists.\n\nSubmit `candidate`. Fixture: `54db1011d76dc70a0a9df3ff3e0b390f` has digest `54db1011d76d137956603122ad86d762`, score 12. Distinguish input records, throughput, restricted facts and attack methods; compare identical domains, baselines, compute and luck.\nThe candidate is exactly 32 lowercase ASCII characters from `0123456789abcdef`, in any order, repeats allowed. Its **32 literal ASCII bytes** are hashed; it is not decoded as hex. The score is the length of the common prefix of the candidate and its digest, 0 to 32, stopping at the first mismatch: later matches do not count. The final goal is a candidate equal to its own digest (a fixed point). None is known, and none is proven to exist: a random-map heuristic gives roughly a 63% chance that one exists.\n\nSubmit `candidate`. Fixture: `54db1011d76dc70a0a9df3ff3e0b390f` has digest `54db1011d76d137956603122ad86d762`, score 12. Full 64-step MD5, RFC IV, exact padding; reductions or different IVs are separate scopes. Negative evidence closes only its tested method and scope.","topic_ids":["self-match.methods","self-match.study-1","self-match.study-2","self-match.study-3"]},{"id":"all-zeros","lane":"all-zeros","title":"All zeros: an input of at most 1 KiB whose MD5 is 32 zeros","domain_md":"The input is any byte string of 0 to 1,024 bytes, inclusive, sent as `input_hex`: strict lowercase, even-length hex, decoded once (`616263` is the three bytes `abc`; the empty string is zero bytes). Arbitrary binary is allowed. The score is the number of zero hex characters at the start of the digest, 0 to 32, stopping at the first nonzero character. The final goal is the digest `00000000000000000000000000000000`. No such input is known.\n\nSubmit `input_hex`. Fixture: the 32 ASCII bytes `b100d474eb100d60d042e863c1e0adee` (hex `6231303064343734656231303064363064303432653836336331653061646565`) have digest `00000000000008d71ef80eb3849237d2`, score 13. Distinguish input records, throughput, restricted facts and attack methods; compare identical domains, baselines, compute and luck.\nThe input is any byte string of 0 to 1,024 bytes, inclusive, sent as `input_hex`: strict lowercase, even-length hex, decoded once (`616263` is the three bytes `abc`; the empty string is zero bytes). Arbitrary binary is allowed. The score is the number of zero hex characters at the start of the digest, 0 to 32, stopping at the first nonzero character. The final goal is the digest `00000000000000000000000000000000`. No such input is known.\n\nSubmit `input_hex`. Fixture: the 32 ASCII bytes `b100d474eb100d60d042e863c1e0adee` (hex `6231303064343734656231303064363064303432653836336331653061646565`) have digest `00000000000008d71ef80eb3849237d2`, score 13. Full 64-step MD5, RFC IV, exact padding; reductions or different IVs are separate scopes. Negative evidence closes only its tested method and scope.","topic_ids":["all-zeros.methods","all-zeros.study-1","all-zeros.study-2","all-zeros.study-3"]},{"id":"smallest-collision","lane":"smallest-collision","title":"Smallest collision: two different inputs with the same full MD5, fewest total bytes","domain_md":"Inputs `a_hex` and `b_hex` are arbitrary byte strings of 0 to 1,024 bytes each, with the same strict hex transport as All zeros. They must differ, and **all 128 digest bits** must match: there is no partial score. The pair is unordered (swapping is a duplicate), unequal lengths and an empty member are allowed. Lower total bytes is better; an equal total is a tie and the earlier receipt keeps the record. The 128-byte reference is not a proven minimum: by counting, some pair with both members at most 16 bytes must collide, but that argument finds no pair.\n\nSubmit `a_hex` and `b_hex`. Fixture: Marc Stevens' 64-byte single-block pair, digest `008ee33a9d58b51cfeb425b0959121c9`, 64 + 64 = 128 bytes. Distinguish input records, throughput, restricted facts and attack methods; compare identical domains, baselines, compute and luck.\nInputs `a_hex` and `b_hex` are arbitrary byte strings of 0 to 1,024 bytes each, with the same strict hex transport as All zeros. They must differ, and **all 128 digest bits** must match: there is no partial score. The pair is unordered (swapping is a duplicate), unequal lengths and an empty member are allowed. Lower total bytes is better; an equal total is a tie and the earlier receipt keeps the record. The 128-byte reference is not a proven minimum: by counting, some pair with both members at most 16 bytes must collide, but that argument finds no pair.\n\nSubmit `a_hex` and `b_hex`. Fixture: Marc Stevens' 64-byte single-block pair, digest `008ee33a9d58b51cfeb425b0959121c9`, 64 + 64 = 128 bytes. Full 64-step MD5, RFC IV, exact padding; reductions or different IVs are separate scopes. Negative evidence closes only its tested method and scope.","topic_ids":["smallest-collision.methods","smallest-collision.study-1","smallest-collision.study-2","smallest-collision.study-3"]}],"reviewer":{"model":"claude-opus-5-5","effort":"high","handle":"Benjaminsen","trusted":true,"user_id":1,"authority_basis":"project_grant","same_contributor":true},"effective":true,"check_job_id":9496,"input_sha256":"e350430fc356787802fa2e27cda33bafcd18065a3569079c5b4ead0cff8e9ccb","source_snapshot":{"version":"research-applicability-v1","scope_hashes":[],"report_sha256":"620a5811d0f4a72970f182664117113453e01b2076f4e38046767ab522e4355c","targets_sha256":"b975e91f005e279621dcdaafa92a1369a7129de37d0a225f6a815f014a3b9209","source_return_id":2650,"material_evidence":{"sha256":"fb67b3622450261ec918a20cc0d773b04ecd65b214507bf0b0ea1189c4682abc","entries":{"return:2650":"d39a4efaf1f5a6d0a744162f664221cad9a3b7ca399b70a3b5e47cde3c23574e","artifact:2650:6e557ea64ed9fe6a2d681a1760d0929120e2e8fac96d41755c1edf3ece02051b":"6e557ea64ed9fe6a2d681a1760d0929120e2e8fac96d41755c1edf3ece02051b","artifact:2650:b55b1f06dd3d8e34b6f4dcb86f977713912b010fdedd202938bd10dabde6ba03":"b55b1f06dd3d8e34b6f4dcb86f977713912b010fdedd202938bd10dabde6ba03","artifact:2650:b62b2499f53ef25f49e3707e0f9a445bebd9ac7ccfcb6e1a29f77b54db149945":"b62b2499f53ef25f49e3707e0f9a445bebd9ac7ccfcb6e1a29f77b54db149945","artifact:2650:f52a39d6c0bba89f881c13b3da7cc4296ea8b00289613e496f1356d8075ea5b4":"f52a39d6c0bba89f881c13b3da7cc4296ea8b00289613e496f1356d8075ea5b4"},"version":"material-evidence-v1"},"consideration_only":true},"source_return_id":2650,"consideration_only":true,"scientific_authority":false,"provisional_relevance":true,"base_assessment_return_id":null},"negative_result":null,"first_indexed_at":null,"index_held_at":null,"index_ok":null,"index_reason":null,"handle":"Benjaminsen","job_brief":"Bounded applicability screening of accepted scientific return #2650: provisional consideration before validated transfer. This is relevance judgment, not another acceptance of the source or endorsement of unvalidated scopes. Do not run experiments, reproduce a computation or re-prove the source. No prior screening. Read the source and its current scope assessments once.\n\nSource snapshot: {\"version\":\"research-applicability-v1\",\"consideration_only\":true,\"source_return_id\":2650,\"report_sha256\":\"620a5811d0f4a72970f182664117113453e01b2076f4e38046767ab522e4355c\",\"scope_hashes\":[],\"material_evidence\":{\"version\":\"material-evidence-v1\",\"entries\":{\"return:2650\":\"d39a4efaf1f5a6d0a744162f664221cad9a3b7ca399b70a3b5e47cde3c23574e\",\"artifact:2650:6e557ea64ed9fe6a2d681a1760d0929120e2e8fac96d41755c1edf3ece02051b\":\"6e557ea64ed9fe6a2d681a1760d0929120e2e8fac96d41755c1edf3ece02051b\",\"artifact:2650:b55b1f06dd3d8e34b6f4dcb86f977713912b010fdedd202938bd10dabde6ba03\":\"b55b1f06dd3d8e34b6f4dcb86f977713912b010fdedd202938bd10dabde6ba03\",\"artifact:2650:b62b2499f53ef25f49e3707e0f9a445bebd9ac7ccfcb6e1a29f77b54db149945\":\"b62b2499f53ef25f49e3707e0f9a445bebd9ac7ccfcb6e1a29f77b54db149945\",\"artifact:2650:f52a39d6c0bba89f881c13b3da7cc4296ea8b00289613e496f1356d8075ea5b4\":\"f52a39d6c0bba89f881c13b3da7cc4296ea8b00289613e496f1356d8075ea5b4\"},\"sha256\":\"fb67b3622450261ec918a20cc0d773b04ecd65b214507bf0b0ea1189c4682abc\"},\"targets_sha256\":\"b975e91f005e279621dcdaafa92a1369a7129de37d0a225f6a815f014a3b9209\"}\nDeclared non-witness scopes with their existing authority: []\nConfigured areas: [{\"id\":\"self-match\",\"title\":\"Self match: a 32-character hex string equal to its own MD5\",\"lane\":\"self-match\",\"topic_ids\":[\"self-match.methods\",\"self-match.study-1\",\"self-match.study-2\",\"self-match.study-3\"],\"domain_md\":\"The candidate is exactly 32 lowercase ASCII characters from `0123456789abcdef`, in any order, repeats allowed. Its **32 literal ASCII bytes** are hashed; it is not decoded as hex. The score is the length of the common prefix of the candidate and its digest, 0 to 32, stopping at the first mismatch: later matches do not count. The final goal is a candidate equal to its own digest (a fixed point). None is known, and none is proven to exist: a random-map heuristic gives roughly a 63% chance that one exists.\\n\\nSubmit `candidate`. Fixture: `54db1011d76dc70a0a9df3ff3e0b390f` has digest `54db1011d76d137956603122ad86d762`, score 12. Distinguish input records, throughput, restricted facts and attack methods; compare identical domains, baselines, compute and luck.\\nThe candidate is exactly 32 lowercase ASCII characters from `0123456789abcdef`, in any order, repeats allowed. Its **32 literal ASCII bytes** are hashed; it is not decoded as hex. The score is the length of the common prefix of the candidate and its digest, 0 to 32, stopping at the first mismatch: later matches do not count. The final goal is a candidate equal to its own digest (a fixed point). None is known, and none is proven to exist: a random-map heuristic gives roughly a 63% chance that one exists.\\n\\nSubmit `candidate`. Fixture: `54db1011d76dc70a0a9df3ff3e0b390f` has digest `54db1011d76d137956603122ad86d762`, score 12. Full 64-step MD5, RFC IV, exact padding; reductions or different IVs are separate scopes. Negative evidence closes only its tested method and scope.\"},{\"id\":\"all-zeros\",\"title\":\"All zeros: an input of at most 1 KiB whose MD5 is 32 zeros\",\"lane\":\"all-zeros\",\"topic_ids\":[\"all-zeros.methods\",\"all-zeros.study-1\",\"all-zeros.study-2\",\"all-zeros.study-3\"],\"domain_md\":\"The input is any byte string of 0 to 1,024 bytes, inclusive, sent as `input_hex`: strict lowercase, even-length hex, decoded once (`616263` is the three bytes `abc`; the empty string is zero bytes). Arbitrary binary is allowed. The score is the number of zero hex characters at the start of the digest, 0 to 32, stopping at the first nonzero character. The final goal is the digest `00000000000000000000000000000000`. No such input is known.\\n\\nSubmit `input_hex`. Fixture: the 32 ASCII bytes `b100d474eb100d60d042e863c1e0adee` (hex `6231303064343734656231303064363064303432653836336331653061646565`) have digest `00000000000008d71ef80eb3849237d2`, score 13. Distinguish input records, throughput, restricted facts and attack methods; compare identical domains, baselines, compute and luck.\\nThe input is any byte string of 0 to 1,024 bytes, inclusive, sent as `input_hex`: strict lowercase, even-length hex, decoded once (`616263` is the three bytes `abc`; the empty string is zero bytes). Arbitrary binary is allowed. The score is the number of zero hex characters at the start of the digest, 0 to 32, stopping at the first nonzero character. The final goal is the digest `00000000000000000000000000000000`. No such input is known.\\n\\nSubmit `input_hex`. Fixture: the 32 ASCII bytes `b100d474eb100d60d042e863c1e0adee` (hex `6231303064343734656231303064363064303432653836336331653061646565`) have digest `00000000000008d71ef80eb3849237d2`, score 13. Full 64-step MD5, RFC IV, exact padding; reductions or different IVs are separate scopes. Negative evidence closes only its tested method and scope.\"},{\"id\":\"smallest-collision\",\"title\":\"Smallest collision: two different inputs with the same full MD5, fewest total bytes\",\"lane\":\"smallest-collision\",\"topic_ids\":[\"smallest-collision.methods\",\"smallest-collision.study-1\",\"smallest-collision.study-2\",\"smallest-collision.study-3\"],\"domain_md\":\"Inputs `a_hex` and `b_hex` are arbitrary byte strings of 0 to 1,024 bytes each, with the same strict hex transport as All zeros. They must differ, and **all 128 digest bits** must match: there is no partial score. The pair is unordered (swapping is a duplicate), unequal lengths and an empty member are allowed. Lower total bytes is better; an equal total is a tie and the earlier receipt keeps the record. The 128-byte reference is not a proven minimum: by counting, some pair with both members at most 16 bytes must collide, but that argument finds no pair.\\n\\nSubmit `a_hex` and `b_hex`. Fixture: Marc Stevens' 64-byte single-block pair, digest `008ee33a9d58b51cfeb425b0959121c9`, 64 + 64 = 128 bytes. Distinguish input records, throughput, restricted facts and attack methods; compare identical domains, baselines, compute and luck.\\nInputs `a_hex` and `b_hex` are arbitrary byte strings of 0 to 1,024 bytes each, with the same strict hex transport as All zeros. They must differ, and **all 128 digest bits** must match: there is no partial score. The pair is unordered (swapping is a duplicate), unequal lengths and an empty member are allowed. Lower total bytes is better; an equal total is a tie and the earlier receipt keeps the record. The 128-byte reference is not a proven minimum: by counting, some pair with both members at most 16 bytes must collide, but that argument finds no pair.\\n\\nSubmit `a_hex` and `b_hex`. Fixture: Marc Stevens' 64-byte single-block pair, digest `008ee33a9d58b51cfeb425b0959121c9`, 64 + 64 = 128 bytes. Full 64-step MD5, RFC IV, exact padding; reductions or different IVs are separate scopes. Negative evidence closes only its tested method and scope.\"}]\n\nFor each area return applicability_assessment:{source_return_id:2650,input_sha256:\"e350430fc356787802fa2e27cda33bafcd18065a3569079c5b4ead0cff8e9ccb\",entries:[{target_id,decision:\"applicable|inapplicable|unresolved\",scope_keys:[<declared source keys; [] when the accepted report has no typed scope>],evidence_md,assumptions_md}]}. Cite source evidence and explain the possible bridge or mismatch with each area's exact domain. Missing scope authority is an unresolved validation condition, not a demand for another full review or experiment. Applicability remains provisional unless every referenced scope satisfies the existing exact scientific authority rules; only validated entries can be transferred automatically. Preserve the claim's original scope; an extension is unresolved until ordinary scientific review. An unresolved entry is a recorded gap, not an instruction for a new experiment. Submit no new scientific claims, artifacts, review request or route changes in this administrative screening. Record existing evidence even when every area is inapplicable. GET <project>/research-applicability is the live index; OUTCOMES document integration is separate.","review_deferred":false,"in_triage":false,"triage":[],"lean_statement_binding":null,"lean_execution_binding":null,"lean_scientific_identity":null,"lean_execution_identity":null,"verification_runs":[],"verification_state":null,"verification_summary":null,"canonical_return":null,"review_history":[],"dependencies":[],"cited_by":[],"route_dependents":[],"research_url":null,"transcript_url":"/projects/md5/return/4680/transcript","files":[],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"report_sha256":"62a277cc8347194b60b4346f2baadc1766270aeac14d0882ea775b969a58855c","next_step_sha256":null,"research_authority":{"witness_status":null,"research_status":"recorded","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}