{"claim":{"id":"002bfc1f-44c6-462e-b006-966cb64c2a0e","submission_id":"ed441aab-f395-4a87-b639-9e0d1a703bdb","submission_revision_id":"50d0334a-b70a-4a8a-96bd-231adfee965c","builder":"builder-01","status":"active","claimed_at":"2026-10-15T15:08:42.195252Z","delivered_at":null,"ended_at":null,"delivery_note":null,"end_reason":null},"task":{"submission_id":"ed441aab-f395-4a87-b639-9e0d1a703bdb","submission_revision_id":"50d0334a-b70a-4a8a-96bd-231adfee965c","revision_version":2,"submission_status":"under_review","submitted_at":"2026-08-20T07:08:41.525946Z","submitter_name":"Hyunjun Kim","submitter_email":"mnb9227@gmail.com","task":{"status":"submitted","version":2,"extra_fields":{"difficulty_self_test":{"model":"Scripted QC baseline (agent run pending)","harness":"manual ffmpeg pipeline, deterministic evaluator","estimated_score":"<1%"}},"submitted_at":"2026-08-20T07:08:41.525946+00:00","industry_domain":"Broadcast & Post-Production (Video)","task_short_name":"broadcast_deliverable_conform","operating_system":"Linux","task_description":"Objective: Produce a spec-compliant broadcast deliverable (video master + conformed subtitles) from raw sources, an edit decision list, and a delivery specification.\n\nStarting State: The task folder contains three 1920x1080 25fps source clips (sources/A.mp4, B.mp4, C.mp4) recorded with DIFFERENT production timecode offsets (camera report: source_tc_map.txt — the clips do not start at TC 00:00:00:00), a CMX3600 non-drop-frame EDL (program.edl; 4 events; source TCs are production timecode; record side starts at 01:00:00:00), subtitles timed to program timecode (subtitles_program.srt), and a complete delivery specification (spec_sheet.md) whose every clause is enforced by QC.\n\nInstructions & Rules:\n1. Map the EDL source timecodes to file positions using the camera report and conform the 4 events frame-accurately (1075 program frames).\n2. Prepend the required 4-second black-and-silent leader (program TC 01:00:00:00 = file time 00:00:04:00; 1175 total frames).\n3. Normalize loudness to EBU R128: integrated -23.0 LUFS (±0.5 LU), true peak <= -1.0 dBTP over the whole file.\n4. Encode to spec: H.264 High profile level 4.0, 1920x1080p25 CFR yuv420p, exactly one IDR every 25 frames, no B-frames, AAC-LC 192 kbps 48 kHz stereo, MP4.\n5. Retime subtitles_program.srt onto the deliverable file timeline (text unchanged) and deliver as output/subtitles.srt.\n\nDeliverables: output/deliverable.mp4 and output/subtitles.srt.\n\nThis mirrors a real post-production conform/QC/delivery workflow: timecode math, loudness compliance, subtitle conforming, and delivery-spec encoding are standard broadcast QC gates, and an error in any stage invalidates the deliverable.","software_licensing":"Free/Open Source","evaluation_criteria":null,"verification_method":"Deterministic QC gates, all measured with ffprobe/ffmpeg (full evaluator provided: evaluation/evaluate.py in the reference package). Hard gates (any failure => 0): - Frame count is exactly 1175 (ffprobe nb_frames). - The 4-second leader is black (sampled frames YAVG <= 16) AND silent: first 3.95 s of audio must peak <= -60 dBFS (volumedetect; 50 ms margin absorbs AAC priming). - True peak <= -1.0 dBTP (loudnorm/ebu128 over the whole file). - Conform accuracy: >= 10 of 12 probe frames (incl. the 3 event cut points) match the hidden reference render at SSIM >= 0.95. - Mandatory delivery spec: MP4 container (ftyp brand isom/iso2/mp41/mp42/avc1); progressive CFR (field_order=progressive + exact 1/25 s packet timestamp grid); AAC-LC with measured bitrate within +/-15% of 192 kbps; actual IDR units (H.264 NAL type-5 count = 1175/25 = 47, parsed from the annex-B bitstream). Scored components (weighted sum after gates pass): - Conform: fraction of the 12 SSIM probes passing (weight 0.4). - Loudness: integrated LUFS within -23.0 +/- 0.5 (weight 0.2). - Subtitles: fraction of 8 cues with exact text and timing within +/-80 ms of the reference SRT (weight 0.2). - Spec conformance: fraction of 14 ffprobe/bitstream checks — h264, High profile, level 4.0, 1920x1080, 25/1 fps, yuv420p, no B-frames, AAC 48 kHz stereo, keyframe packet every 25 frames, MP4 container, progressive CFR, AAC-LC, 192 kbps +/-15%, exact IDR count (weight 0.2). Sanity checks retained in the reference package (selftest_review_log.txt): the reference render itself scores 1.0; the recorded naive attempt scores 0.0; targeted violations — audible leader audio, keyint-50/scene-cut non-IDR GOPs, and MP3 44.1 kHz mono audio — are each rejected at 0.0.","software_and_version":"ffmpeg 9.0 (ffprobe, loudnorm/EBU R128, libx264)","specific_requirements":null,"expected_completion_time":null},"files":[{"file_id":"75ac0e01-0551-46a8-944c-5ac194b098de","filename":"upload_inputs.zip","category":"input_materials","content_type":"application/x-zip-compressed","size_bytes":100481727},{"file_id":"53e53db4-813c-4efb-a43a-4044a19bbaec","filename":"upload_reference.zip","category":"reference_output","content_type":"application/x-zip-compressed","size_bytes":56122321}],"deep_review":{"overall_verdict":"Strong Accept","review_text":"**Overall Verdict:** Strong Accept — The submission is complete, internally consistent, professionally representative, and objectively evaluable end to end with no material contributor-side gaps.\n\n**Completeness Check:**\n- Instruction: Complete — The conform, leader, loudness, encoding, subtitle, frame-count, and output requirements are explicit and mutually consistent.\n- Inputs: Complete — Three valid 1080p25 AVC/AAC sources, the four-event EDL, production-timecode map, delivery specification, and eight-cue program SRT are present; all EDL ranges fit the source durations and total 1,075 program frames.\n- Toolchain: Complete — FFmpeg/ffprobe with loudnorm and libx264 is legally obtainable on Linux, requires no specialized hardware, and supports the entire workflow on one ordinary machine.\n- Reference output: Complete — The MP4/SRT pair is present. Direct container and bitstream inspection confirms 47 seconds, 1,175 uniform 1080p25 AVC samples, High profile level 4.0 headers, 47 type-5 IDRs at 25-frame cadence, no composition offsets, and 48 kHz stereo AAC. The eight subtitle cues have the correct four-second shift and unchanged text.\n- Rubric: Complete — The readable evaluator defines deterministic hard gates, weighted partial scoring, tolerances, reference comparisons, and negative self-tests that distinguish compliant, naive, and targeted-invalid outputs.\n\n**File and Task Reachability:**\nAttachment classification: `partially_reachable`. Task classification: `reachable`. Both manifest ZIPs matched their declared byte counts, passed integrity testing, were unencrypted, and contained only regular files without traversal or unsafe links. Inspected members were `input/program.edl`, `input/source_tc_map.txt`, `input/spec_sheet.md`, `input/subtitles_program.srt`, `input/sources/A.mp4`, `B.mp4`, and `C.mp4`; `reference/deliverable.mp4`, `reference/subtitles.srt`, `evaluation/evaluate.py`, `naive_try/deliverable.mp4`, `naive_try/subtitles.srt`, `task_card.json`, `build_log.txt`, and `selftest_review_log.txt`. No submitted program was executed. Decoded picture inspection and independent black, silence, SSIM, and loudness measurements could not be repeated because the reviewer environment lacked FFmpeg/ffprobe; container tables, codec headers, all AVC samples, logs, evaluator source, and self-test records were inspected instead. This is a reviewer-tooling limitation, not a submission defect. The end-to-end path is clear: map EDL timecodes to source frames, trim and concatenate the four events behind the leader, normalize audio, encode the required MP4, retime subtitles, and evaluate both outputs through deterministic media and reference checks. License keys, accounts, tokens, and other credentials are not required from the contributor. Confidential, proprietary, or internal material may be evaluated as submitted evidence without requiring additional disclosure.\n\n**Special Screening:**\nNone\n\n**Recommended Changes:**\nThe timecode construction, reference pair, evaluator, and targeted negative tests provide sufficient evidence for deployment. No changes are required.","reviewed_at":"2026-08-20T07:35:32.067333Z"},"claim":{"id":"002bfc1f-44c6-462e-b006-966cb64c2a0e","submission_id":"ed441aab-f395-4a87-b639-9e0d1a703bdb","submission_revision_id":"50d0334a-b70a-4a8a-96bd-231adfee965c","builder":"builder-01","status":"active","claimed_at":"2026-10-15T15:08:42.195252Z","delivered_at":null,"ended_at":null,"delivery_note":null,"end_reason":null}},"revision_is_current":true}