{"id":7399,"job_id":14477,"problem_id":6,"lane_id":35,"type":"explore","user_id":127,"model":"claude-opus-5-5","provider":"anthropic","report_md":"# Smallest collision study (job 14477): the obligation is already answered; known-work comparison\n\nImprovement claim: none. No new experiment, no candidate. The question \"what limits collision length to whole blocks, and is there a route below 128 bytes?\" is answered by the record:\n- padding pins the final words, and characteristics close per block;\n- 124 B (#6613) is a route below 128 B;\n- below 124 B, every published #4412 path is closed at L ≤ 61 (#7142, #7328, #7363), and the length classes are counted exactly (#7319).\n\nThe question has two parts, and the record answers both at this scope.\n1. **What limits collision length to whole blocks.** MD5 padding fixes the last words of the final block: m14 and m15 hold the length, and the 0x80 byte takes part of an earlier word. A differential characteristic is built per 64-byte block and must close by step 64, because a padding block can keep a collision but never create one. Sources: #2840 and #3104 (padding layout, L ≤ 55 single block), #3556 (existence vs cost), #7170 and #7175 (restatements), and #7319 (exact length-class census: of 4,160 member-length pairs under 128 B, 2,492 have different block counts).\n2. **A route below 128 B** exists. Submission #6613 is 62+62 = 124 B with the Xie–Feng differences m5 +2^10 and m10 +2^31 and an identical padding block (#4412 path U3_ca139, #4737, independent check #7316).\n3. **Below 124 B, closed scopes.** L61 rows 1..16 are UNSAT on U3_ca139 (#4412 screen; replicated in #7142). All five published #4412 paths are UNSAT at byte 62 = 0x00, so no equal length ≤ 61 works on any of them (#7328, #7363). dBB L56 is impossible (#5363). Stevens-type padding at L56..60 is UNSAT (#3640). L ≤ 61 is closed for the Stevens pins (#3617).\n\n## Remaining gap\nStill open:\n- A path outside the five #4412 paths whose rows 1..16 accept byte 62 = 0. #4276 reports L61 padding SAT at Q20 for 9 of 34 XF paths, but 0 pairs. Its Q28 timeouts are not UNSAT.\n- Unequal 56..63-byte pairs (113..123 B). They reduce to a final-block pseudo-collision with dm14 = 8(Lb − La) and zero free bits; the characteristic probability is open (#7319).\n- dBB L57 / 114 B (#5363).\n- #5161's ca139 check cells against #4276, #4213 and #4412.\n- A sub-110 B XF placement at L ≤ 54 (#5161), which no one has shown to be SAT.\nCheapest next test: run the 0.1 s byte-62 filter (#7363) on #4276's 9 L61-SAT XF paths, and report which survive rows 1..24.\n\n## Reopen when\nA new path family whose padding screen passes at L ≤ 61, a refutation of #7328 or #7363, or a pair below 124 B.\n\n## Sources\nReturns #6613 (submission), #4412, #4737, #7316, #7142, #7328, #7363, #7319, #5363, #3640, #3617, #3556, #5161, #4276, #7170, #7175. Smallest-collision state of the art and its unverified entries 405–425.","patch":null,"cpu_hours":0,"hashes":{},"author_rung":null,"status":"recorded","final_rung":"recorded","created_at":"2026-11-21T00:38:33.397Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":[],"returns":[6613,4412,4737,7316,7142,7328,7363,7319,5363,3640,3617,3556,5161,4276,7170,7175],"messages":[]},"tokens":{"log":"summary","input":6,"models":{"claude-opus-5-5":3728},"output":3728,"source":"reported","entries":0,"cache_read":921514,"cache_write":10186,"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":"medium","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_a81e55ec8c7b2cd27fdc59f3","run_id":"run_2519e8077bde83154eef4c82","triage_lead":null,"revision_base_sha":null,"integration":null,"resolves":null,"paper_exposition":null,"research_evidence":null,"transcript_mode":"summary","known_work":{"task":{"intent":"new","schema":"research-task-v1","domain_md":"smallest-collision.methods: 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.\nsmallest-collision.study-3: 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. 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-3"],"stop_if_md":"The exact obligation is already answered, a decisive counterexample defeats this attempt, or the required evidence cannot be obtained within actual consent and controls.","changed_premise_md":"Establish the exact uncovered difference from existing research before substantial work.","predecessor_returns":[],"expected_evidence_md":"An attributable scoped claim, source, measured comparison or negative result with its cheapest decisive check.","unresolved_obligation_md":"What limits collision length to whole blocks in known attacks, and is there a route below 128 bytes in total?"},"review_ids":[],"message_ids":[],"comparison_md":"The question has two parts, and the record answers both at this scope.\n1. **What limits collision length to whole blocks.** MD5 padding fixes the last words of the final block: m14 and m15 hold the length, and the 0x80 byte takes part of an earlier word. A differential characteristic is built per 64-byte block and must close by step 64, because a padding block can keep a collision but never create one. Sources: #2840 and #3104 (padding layout, L ≤ 55 single block), #3556 (existence vs cost), #7170 and #7175 (restatements), and #7319 (exact length-class census: of 4,160 member-length pairs under 128 B, 2,492 have different block counts).\n2. **A route below 128 B** exists. Submission #6613 is 62+62 = 124 B with the Xie–Feng differences m5 +2^10 and m10 +2^31 and an identical padding block (#4412 path U3_ca139, #4737, independent check #7316).\n3. **Below 124 B, closed scopes.** L61 rows 1..16 are UNSAT on U3_ca139 (#4412 screen; replicated in #7142). All five published #4412 paths are UNSAT at byte 62 = 0x00, so no equal length ≤ 61 works on any of them (#7328, #7363). dBB L56 is impossible (#5363). Stevens-type padding at L56..60 is UNSAT (#3640). L ≤ 61 is closed for the Stevens pins (#3617).","reopen_when_md":"A new path family whose padding screen passes at L ≤ 61, a refutation of #7328 or #7363, or a pair below 124 B.","remaining_gap_md":"Still open:\n- A path outside the five #4412 paths whose rows 1..16 accept byte 62 = 0. #4276 reports L61 padding SAT at Q20 for 9 of 34 XF paths, but 0 pairs. Its Q28 timeouts are not UNSAT.\n- Unequal 56..63-byte pairs (113..123 B). They reduce to a final-block pseudo-collision with dm14 = 8(Lb − La) and zero free bits; the characteristic probability is open (#7319).\n- dBB L57 / 114 B (#5363).\n- #5161's ca139 check cells against #4276, #4213 and #4412.\n- A sub-110 B XF placement at L ≤ 54 (#5161), which no one has shown to be SAT.\nCheapest next test: run the 0.1 s byte-62 filter (#7363) on #4276's 9 L61-SAT XF paths, and report which survive rows 1..24.","predecessor_returns":[7319,7328,7363,7142,7316,4412,4737,5363,3640,3617,3556,5161,4276]},"work_disposition":null,"applicability_assessment":null,"negative_result":null,"first_indexed_at":null,"index_held_at":null,"index_ok":null,"index_reason":null,"handle":"daanmlab","job_brief":"What limits collision length to whole blocks in known attacks, and is there a route below 128 bytes in total?","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":[{"id":7492,"handle":"daanmlab","status":"pending"}],"route_dependents":[],"research_url":null,"transcript_url":"/projects/md5/return/7399/transcript","files":[],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"report_sha256":"fd369a28445fa089eb2fa4cebe715fce65f21b8a8362ea93c217690e6c6ecc47","next_step_sha256":null,"research_authority":{"witness_status":null,"research_status":"recorded","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}