TPBench measures three retrieval targets after dialogue compression: This repository includes data, construction inputs, compression and reader code, scorers, saved answers, and paper results.
Dataset Card
TPBench measures three retrieval targets after dialogue compression: This repository includes data, construction inputs, compression and reader code, scorers, saved answers, and paper results. RESULTSMAP.md connects the paper comparisons to their evidence files. Use Python 3.10 or newer. No model access is needed for these checks. The checks validate file hashes and metadata, regenerate all 42 main probe cells, rescore saved answers, and test FMTS context construction. Outputs from tests are written to a temporary directory, not over the released results. Each row fixes the example identifier, method, budget, reference answer, and reader prompts. Identifiers remain stable across…
Excerpt from the card by Papersubmission, licensed other.
Structure
current_value 22,932 rows
| Split | Rows | Size |
|---|---|---|
| probes | 22,932 | 48.6 MB |
initial_goal 19,200 rows
| Split | Rows | Size |
|---|---|---|
| probes | 19,200 | 33.0 MB |
joint_recovery 9,600 rows
| Split | Rows | Size |
|---|---|---|
| probes | 9,600 | 21.0 MB |
Details
- Repository
- 4papersubmission/TPBench
- Publisher
- Papersubmission
- Task category
- Question answering
- Tags
- dialogue, long-context, evaluation
- Size category
- Not stated by the source
- Languages
- en, zh
- Revision
- 4cf58bb41539094bd394c9abd84dea95c0396340
- Last updated
- 2026-09-28
Files
476 files, 572.5 MB in total.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| LICENSES/AnswerEquivalence-Apache-2.0.txt | Data | 11.4 KB | — |
| LICENSES/Apache-2.0.txt | Data | 11.4 KB | — |
| LICENSES/CC-BY-4.0.txt | Data | 547 B | — |
| LICENSES/LongMemEval.txt | Data | 1.0 KB | — |
| LICENSES/MultiWOZ_2.2.txt | Data | 582 B | — |
| LICENSES/RiSAWOZ.txt | Data | 1.3 KB | — |
| LICENSES/SGD.txt | Data | 765 B | — |
| batches/fmts_refpool/gold_tqa_indirect_seed42_refpool_r010.jsonl | Data | 508.4 KB | — |
| batches/fmts_refpool/gold_tqa_indirect_seed42_refpool_r030.jsonl | Data | 508.4 KB | — |
| batches/fmts_refpool/gold_tqa_multiwoz_scope_refpool_r010.jsonl | Data | 697.3 KB | — |
| batches/fmts_refpool/gold_tqa_multiwoz_scope_refpool_r030.jsonl | Data | 697.3 KB | — |
| batches/fmts_refpool/gold_tqa_sgd_scope_refpool_r010.jsonl | Data | 721.7 KB | — |
| batches/fmts_refpool/gold_tqa_sgd_scope_refpool_r030.jsonl | Data | 721.7 KB | — |
| batches/fmts_refpool/responses_tqa_indirect_seed42_refpool_r010.jsonl | Data | 56.6 KB | — |
| batches/fmts_refpool/responses_tqa_indirect_seed42_refpool_r030.jsonl | Data | 58.3 KB | — |
| batches/fmts_refpool/responses_tqa_multiwoz_scope_refpool_r010.jsonl | Data | 100.6 KB | — |
| batches/fmts_refpool/responses_tqa_multiwoz_scope_refpool_r030.jsonl | Data | 101.4 KB | — |
| batches/fmts_refpool/responses_tqa_sgd_scope_refpool_r010.jsonl | Data | 90.8 KB | — |
| batches/fmts_refpool/responses_tqa_sgd_scope_refpool_r030.jsonl | Data | 92.4 KB | — |
| croissant.json | Data | 45.8 KB | — |
| data/construction/multiwoz_compressed_r10.jsonl | Data | 2.0 MB | — |
| data/construction/multiwoz_compressed_r30.jsonl | Data | 3.3 MB | — |
| data/construction/multiwoz_compressed_r50.jsonl | Data | 3.3 MB | — |
| data/construction/multiwoz_compressed_r70.jsonl | Data | 4.2 MB | — |
| data/construction/sgd_compressed_r10.jsonl | Data | 2.8 MB | — |
| data/construction/sgd_compressed_r30.jsonl | Data | 4.5 MB | — |
| data/construction/sgd_compressed_r50.jsonl | Data | 4.4 MB | — |
| data/construction/sgd_compressed_r70.jsonl | Data | 5.7 MB | — |
| data/fmts/fmts_indirect_seed42.jsonl | Data | 327.0 KB | — |
| data/fmts/fmts_multiwoz_scope.jsonl | Data | 489.6 KB | — |
| data/fmts/fmts_sgd_scope.jsonl | Data | 539.4 KB | — |
| data/probes/probes_lme_ku_all72_r30_s42-46.jsonl | Data | 50.5 MB | 4cff2db83727 |
| data/probes/probes_lme_ku_s42_r10_n72.jsonl | Data | 20.9 MB | a2e17b0b2c31 |
| data/probes/probes_lme_ku_s42_r30_n72.jsonl | Data | 32.6 MB | 61ac40f7c431 |
| data/probes/probes_multiwoz_s42_r10_p1_n200.jsonl | Data | 3.1 MB | — |
| data/probes/probes_multiwoz_s42_r10_p3_n200.jsonl | Data | 2.4 MB | — |
| data/probes/probes_multiwoz_s42_r30_n200_late_intent.jsonl | Data | 4.4 MB | — |
| data/probes/probes_multiwoz_s42_r30_p1_n200.jsonl | Data | 3.9 MB | — |
| data/probes/probes_multiwoz_s42_r30_p3_n200.jsonl | Data | 3.5 MB | — |
| data/probes/probes_multiwoz_s42_r50_p3_n200.jsonl | Data | 2.3 MB | — |
| data/probes/probes_multiwoz_s42_r70_p3_n200.jsonl | Data | 2.7 MB | — |
| data/probes/probes_multiwoz_s43_r10_p1_n200.jsonl | Data | 3.1 MB | — |
| data/probes/probes_multiwoz_s43_r10_p3_n200.jsonl | Data | 2.5 MB | — |
| data/probes/probes_multiwoz_s43_r30_n200_late_intent.jsonl | Data | 4.4 MB | — |
| data/probes/probes_multiwoz_s43_r30_p1_n200.jsonl | Data | 3.9 MB | — |
| data/probes/probes_multiwoz_s43_r30_p3_n200.jsonl | Data | 3.2 MB | — |
| data/probes/probes_multiwoz_s43_r50_p3_n200.jsonl | Data | 2.4 MB | — |
| data/probes/probes_multiwoz_s43_r70_p3_n200.jsonl | Data | 2.9 MB | — |
| data/probes/probes_multiwoz_s44_r10_p1_n200.jsonl | Data | 3.1 MB | — |
| data/probes/probes_multiwoz_s44_r10_p3_n200.jsonl | Data | 2.5 MB | — |
| data/probes/probes_multiwoz_s44_r30_n200_late_intent.jsonl | Data | 4.3 MB | — |
| data/probes/probes_multiwoz_s44_r30_p1_n200.jsonl | Data | 3.8 MB | — |
| data/probes/probes_multiwoz_s44_r30_p3_n200.jsonl | Data | 3.1 MB | — |
| data/probes/probes_multiwoz_s44_r50_p3_n200.jsonl | Data | 2.3 MB | — |
| data/probes/probes_multiwoz_s44_r70_p3_n200.jsonl | Data | 2.8 MB | — |
| data/probes/probes_risawoz_all207_r30_s42-46.jsonl | Data | 7.8 MB | — |
| data/probes/probes_risawoz_s42_r10_n200.jsonl | Data | 3.7 MB | — |
| data/probes/probes_risawoz_s42_r30_n200.jsonl | Data | 4.5 MB | — |
| data/probes/probes_sgd_s42_r10_p1_n200.jsonl | Data | 3.1 MB | — |
| data/probes/probes_sgd_s42_r10_p3_n200.jsonl | Data | 1.9 MB | — |
| data/probes/probes_sgd_s42_r30_n200_late_intent.jsonl | Data | 4.1 MB | — |
| data/probes/probes_sgd_s42_r30_p1_n200.jsonl | Data | 4.0 MB | — |
| data/probes/probes_sgd_s42_r30_p3_n200.jsonl | Data | 2.8 MB | — |
| data/probes/probes_sgd_s42_r50_p3_n200.jsonl | Data | 1.9 MB | — |
| data/probes/probes_sgd_s42_r70_p3_n200.jsonl | Data | 2.2 MB | — |
| data/probes/probes_sgd_s43_r10_p1_n200.jsonl | Data | 3.1 MB | — |
| data/probes/probes_sgd_s43_r10_p3_n200.jsonl | Data | 1.8 MB | — |
| data/probes/probes_sgd_s43_r30_n200_late_intent.jsonl | Data | 4.2 MB | — |
| data/probes/probes_sgd_s43_r30_p1_n200.jsonl | Data | 3.9 MB | — |
| data/probes/probes_sgd_s43_r30_p3_n200.jsonl | Data | 2.3 MB | — |
| data/probes/probes_sgd_s43_r50_p3_n200.jsonl | Data | 1.8 MB | — |
| data/probes/probes_sgd_s43_r70_p3_n200.jsonl | Data | 2.1 MB | — |
| data/probes/probes_sgd_s44_r10_p1_n200.jsonl | Data | 3.1 MB | — |
| data/probes/probes_sgd_s44_r10_p3_n200.jsonl | Data | 2.0 MB | — |
| data/probes/probes_sgd_s44_r30_n200_late_intent.jsonl | Data | 4.2 MB | — |
| data/probes/probes_sgd_s44_r30_p1_n200.jsonl | Data | 3.9 MB | — |
| data/probes/probes_sgd_s44_r30_p3_n200.jsonl | Data | 2.5 MB | — |
| data/probes/probes_sgd_s44_r50_p3_n200.jsonl | Data | 1.9 MB | — |
| data/probes/probes_sgd_s44_r70_p3_n200.jsonl | Data | 2.3 MB | — |
| data/semantic_evaluation/ae_development_inputs.jsonl | Data | 212.6 KB | — |
| data/semantic_evaluation/ae_development_key.jsonl | Data | 75.2 KB | — |
| data/semantic_evaluation/ae_source_manifest.json | Data | 4.3 KB | — |
| data/semantic_evaluation/ae_test_inputs.jsonl | Data | 582.6 KB | — |
| data/semantic_evaluation/ae_test_key.jsonl | Data | 225.7 KB | — |
| data/semantic_evaluation/core_goal_prompt.txt | Data | 2.4 KB | — |
| data/semantic_evaluation/judge_prompt.txt | Data | 2.1 KB | — |
| data/semantic_evaluation/p1_inputs.jsonl | Data | 650.1 KB | — |
| data/semantic_evaluation/p1_key.jsonl | Data | 2.8 MB | — |
| data/semantic_evaluation/p1_source_manifest.json | Data | 1.1 KB | — |
| data/semantic_evaluation/rubric_controls_inputs.jsonl | Data | 3.1 KB | — |
| data/sources/longmemeval_knowledge_updates.json | Data | 40.1 MB | 3b376e0c0d1f |
| data/sources/risawoz_test.json | Data | 11.4 MB | 39aad40f5357 |
| dataset_card/croissant.json | Data | 45.8 KB | — |
| manifest.json | Data | 60.1 KB | — |
| requirements-fmts.txt | Data | 171 B | — |
| requirements.txt | Data | 1.0 KB | — |
| results/aggregates_flat.jsonl | Data | 153.1 KB | — |
| results/bench/efficiency_multiwoz_r10.json | Data | 1.2 KB | — |
| results/bench/efficiency_multiwoz_r30.json | Data | 1.8 KB | — |
| results/bench/efficiency_sgd_r10.json | Data | 1.2 KB | — |
| results/bench/efficiency_sgd_r30.json | Data | 2.1 KB | — |
| results/diagnostics/chunkkv_chunklen/mw_r30_p3_chunkkv_chunklen_r03_aggregate.json | Data | 880 B | — |
| results/diagnostics/chunkkv_chunklen/mw_r30_p3_chunkkv_chunklen_r03_scored.jsonl | Data | 491.0 KB | — |
| results/diagnostics/mistral_128/mw_r10_s42_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r10_s42_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r10_s42_p3_reader_all.jsonl | Data | 1.2 MB | — |
| results/diagnostics/mistral_128/mw_r10_s42_p3_scored.jsonl | Data | 970.6 KB | — |
| results/diagnostics/mistral_128/mw_r10_s43_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r10_s43_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r10_s43_p3_reader_all.jsonl | Data | 1.3 MB | — |
| results/diagnostics/mistral_128/mw_r10_s43_p3_scored.jsonl | Data | 1.0 MB | — |
| results/diagnostics/mistral_128/mw_r10_s44_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r10_s44_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r10_s44_p3_reader_all.jsonl | Data | 1.3 MB | — |
| results/diagnostics/mistral_128/mw_r10_s44_p3_scored.jsonl | Data | 1.0 MB | — |
| results/diagnostics/mistral_128/mw_r30_s42_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r30_s42_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r30_s42_p3_reader_all.jsonl | Data | 1.7 MB | — |
| results/diagnostics/mistral_128/mw_r30_s42_p3_scored.jsonl | Data | 1.4 MB | — |
| results/diagnostics/mistral_128/mw_r30_s43_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r30_s43_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r30_s43_p3_reader_all.jsonl | Data | 1.3 MB | — |
| results/diagnostics/mistral_128/mw_r30_s43_p3_scored.jsonl | Data | 1.0 MB | — |
| results/diagnostics/mistral_128/mw_r30_s44_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r30_s44_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/mw_r30_s44_p3_reader_all.jsonl | Data | 1.3 MB | — |
| results/diagnostics/mistral_128/mw_r30_s44_p3_scored.jsonl | Data | 1.0 MB | — |
| results/diagnostics/mistral_128/seed_summary_mistral128.json | Data | 127.0 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s42_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s42_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s42_p3_reader_all.jsonl | Data | 935.4 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s42_p3_scored.jsonl | Data | 763.0 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s43_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s43_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s43_p3_reader_all.jsonl | Data | 914.0 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s43_p3_scored.jsonl | Data | 744.1 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s44_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s44_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s44_p3_reader_all.jsonl | Data | 965.8 KB | — |
| results/diagnostics/mistral_128/sgd_r10_s44_p3_scored.jsonl | Data | 784.1 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s42_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s42_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s42_p3_reader_all.jsonl | Data | 1.3 MB | — |
| results/diagnostics/mistral_128/sgd_r30_s42_p3_scored.jsonl | Data | 1.1 MB | — |
| results/diagnostics/mistral_128/sgd_r30_s43_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s43_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s43_p3_reader_all.jsonl | Data | 918.0 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s43_p3_scored.jsonl | Data | 741.3 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s44_p1_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s44_p3_aggregate.json | Data | 2.1 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s44_p3_reader_all.jsonl | Data | 971.0 KB | — |
| results/diagnostics/mistral_128/sgd_r30_s44_p3_scored.jsonl | Data | 781.3 KB | — |
| results/diagnostics/update_evidence/fmts_support_ablation_input.jsonl | Data | 712.9 KB | — |
| results/diagnostics/update_evidence/fmts_support_ablation_reader_llama.jsonl | Data | 133.2 KB | — |
| results/diagnostics/update_evidence/fmts_support_ablation_scored_llama.jsonl | Data | 134.9 KB | — |
| results/diagnostics/update_evidence/fmts_support_ablation_summary_llama.json | Data | 1.6 KB | — |
| results/diagnostics/update_evidence/multiwoz_filtered_gold.jsonl | Data | 1.2 MB | — |
| results/diagnostics/update_evidence/multiwoz_goal_labels.jsonl | Data | 749.0 KB | — |
| results/diagnostics/update_evidence/origin_index_multiwoz.jsonl | Data | 327.7 KB | — |
| results/diagnostics/update_evidence/origin_index_sgd.jsonl | Data | 403.8 KB | — |
| results/diagnostics/update_evidence/origin_summary.json | Data | 2.0 KB | — |
| results/diagnostics/update_evidence/p2_aggregate_llama.json | Data | 2.8 KB | — |
| results/diagnostics/update_evidence/p2_all_reader_llama.jsonl | Data | 1.7 MB | — |
| results/diagnostics/update_evidence/p2_reader_input.jsonl | Data | 6.6 MB | — |
| results/diagnostics/update_evidence/p2_reader_input_multiwoz.jsonl | Data | 3.6 MB | — |
| results/diagnostics/update_evidence/p2_reader_input_sgd.jsonl | Data | 3.0 MB | — |
| results/diagnostics/update_evidence/p2_scored_llama.jsonl | Data | 1.7 MB | — |
| results/diagnostics/update_evidence/p2_summary_llama.json | Data | 8.2 KB | — |
| results/diagnostics/update_evidence/p2_variants_multiwoz.jsonl | Data | 2.9 MB | — |
| results/diagnostics/update_evidence/p2_variants_sgd.jsonl | Data | 2.4 MB | — |
| results/diagnostics/update_evidence/p3_crossseed_llama.json | Data | 54.7 KB | — |
| results/diagnostics/update_evidence/p3_crossseed_mistral.json | Data | 55.1 KB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s42_aggregate_chunkkv.json | Data | 405 B | — |
| results/diagnostics/update_evidence/p3_multiwoz_s42_aggregate_llama.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s42_aggregate_mistral.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s42_reader_chunkkv.jsonl | Data | 201.2 KB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s42_reader_llama.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s42_reader_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s42_scored_chunkkv.jsonl | Data | 214.3 KB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s42_scored_llama.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s42_scored_mistral.jsonl | Data | 1.6 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s43_aggregate_llama.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s43_aggregate_mistral.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s43_reader_llama.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s43_reader_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s43_scored_llama.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s43_scored_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s44_aggregate_llama.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s44_aggregate_mistral.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s44_reader_llama.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s44_reader_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s44_scored_llama.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_multiwoz_s44_scored_mistral.jsonl | Data | 1.6 MB | — |
| results/diagnostics/update_evidence/p3_pool_multiwoz.json | Data | 11.3 KB | — |
| results/diagnostics/update_evidence/p3_pool_sgd.json | Data | 14.1 KB | — |
| results/diagnostics/update_evidence/p3_sgd_s42_aggregate_chunkkv.json | Data | 412 B | — |
| results/diagnostics/update_evidence/p3_sgd_s42_aggregate_llama.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_sgd_s42_aggregate_mistral.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_sgd_s42_reader_chunkkv.jsonl | Data | 197.8 KB | — |
| results/diagnostics/update_evidence/p3_sgd_s42_reader_llama.jsonl | Data | 1.4 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s42_reader_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s42_scored_chunkkv.jsonl | Data | 205.9 KB | — |
| results/diagnostics/update_evidence/p3_sgd_s42_scored_llama.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s42_scored_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s43_aggregate_llama.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_sgd_s43_aggregate_mistral.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_sgd_s43_reader_llama.jsonl | Data | 1.4 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s43_reader_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s43_scored_llama.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s43_scored_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s44_aggregate_llama.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_sgd_s44_aggregate_mistral.json | Data | 3.2 KB | — |
| results/diagnostics/update_evidence/p3_sgd_s44_reader_llama.jsonl | Data | 1.4 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s44_reader_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s44_scored_llama.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/p3_sgd_s44_scored_mistral.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/per_item_multiwoz.jsonl | Data | 108.3 KB | — |
| results/diagnostics/update_evidence/per_item_sgd.jsonl | Data | 84.2 KB | — |
| results/diagnostics/update_evidence/probes_multiwoz_s42_r30_n200_late_intent.jsonl | Data | 4.4 MB | — |
| results/diagnostics/update_evidence/probes_multiwoz_s42_r30_n200_late_intent_chunkkv.jsonl | Data | 980.6 KB | — |
| results/diagnostics/update_evidence/probes_multiwoz_s43_r30_n200_late_intent.jsonl | Data | 4.4 MB | — |
| results/diagnostics/update_evidence/probes_multiwoz_s44_r30_n200_late_intent.jsonl | Data | 4.3 MB | — |
| results/diagnostics/update_evidence/probes_sgd_s42_r30_n200_late_intent.jsonl | Data | 4.1 MB | — |
| results/diagnostics/update_evidence/probes_sgd_s42_r30_n200_late_intent_chunkkv.jsonl | Data | 927.4 KB | — |
| results/diagnostics/update_evidence/probes_sgd_s43_r30_n200_late_intent.jsonl | Data | 4.2 MB | — |
| results/diagnostics/update_evidence/probes_sgd_s44_r30_n200_late_intent.jsonl | Data | 4.2 MB | — |
| results/diagnostics/update_evidence/rank_inconsistency_final_slot.json | Data | 10.4 KB | — |
| results/diagnostics/update_evidence/sgd_filtered_gold.jsonl | Data | 1.5 MB | — |
| results/diagnostics/update_evidence/sgd_goal_labels.jsonl | Data | 924.6 KB | — |
| results/fmts_execution/answers.jsonl | Data | 2.0 MB | — |
| results/fmts_execution/protocol.json | Data | 1.1 KB | — |
| results/fmts_execution/tasks.jsonl | Data | 4.0 MB | — |
| results/fmts_refpool/indirect_seed42_refpool_r010.json | Data | 1.7 MB | — |
| results/fmts_refpool/indirect_seed42_refpool_r030.json | Data | 2.2 MB | — |
| results/fmts_refpool/multiwoz_scope_refpool_r010.json | Data | 2.8 MB | — |
| results/fmts_refpool/multiwoz_scope_refpool_r030.json | Data | 3.5 MB | — |
| results/fmts_refpool/scored_tqa_indirect_seed42_refpool_r010.json | Data | 98.0 KB | — |
| results/fmts_refpool/scored_tqa_indirect_seed42_refpool_r030.json | Data | 99.4 KB | — |
| results/fmts_refpool/scored_tqa_multiwoz_scope_refpool_r010.json | Data | 182.4 KB | — |
| results/fmts_refpool/scored_tqa_multiwoz_scope_refpool_r030.json | Data | 183.0 KB | — |
| results/fmts_refpool/scored_tqa_sgd_scope_refpool_r010.json | Data | 172.6 KB | — |
| results/fmts_refpool/scored_tqa_sgd_scope_refpool_r030.json | Data | 173.9 KB | — |
| results/fmts_refpool/sgd_scope_refpool_r010.json | Data | 2.9 MB | — |
| results/fmts_refpool/sgd_scope_refpool_r030.json | Data | 3.7 MB | — |
| results/kv_methods/multiwoz_r30_s42_p1_pyramidkv_aggregate.json | Data | 269 B | — |
| results/kv_methods/multiwoz_r30_s42_p1_snapkv_aggregate.json | Data | 266 B | — |
| results/kv_methods/multiwoz_r30_s42_p1_streamingllm_aggregate.json | Data | 276 B | — |
| results/kv_methods/multiwoz_r30_s42_p3_pyramidkv_aggregate.json | Data | 309 B | — |
| results/kv_methods/multiwoz_r30_s42_p3_pyramidkv_scored.jsonl | Data | 166.3 KB | — |
| results/kv_methods/multiwoz_r30_s42_p3_snapkv_aggregate.json | Data | 307 B | — |
| results/kv_methods/multiwoz_r30_s42_p3_snapkv_scored.jsonl | Data | 165.8 KB | — |
| results/kv_methods/multiwoz_r30_s42_p3_streamingllm_aggregate.json | Data | 314 B | — |
| results/kv_methods/multiwoz_r30_s42_p3_streamingllm_scored.jsonl | Data | 167.0 KB | — |
| results/kv_methods/sgd_r30_s42_p1_pyramidkv_aggregate.json | Data | 272 B | — |
| results/kv_methods/sgd_r30_s42_p1_snapkv_aggregate.json | Data | 264 B | — |
| results/kv_methods/sgd_r30_s42_p1_streamingllm_aggregate.json | Data | 271 B | — |
| results/kv_methods/sgd_r30_s42_p3_pyramidkv_aggregate.json | Data | 303 B | — |
| results/kv_methods/sgd_r30_s42_p3_pyramidkv_scored.jsonl | Data | 131.0 KB | — |
| results/kv_methods/sgd_r30_s42_p3_snapkv_aggregate.json | Data | 300 B | — |
| results/kv_methods/sgd_r30_s42_p3_snapkv_scored.jsonl | Data | 130.5 KB | — |
| results/kv_methods/sgd_r30_s42_p3_streamingllm_aggregate.json | Data | 302 B | — |
| results/kv_methods/sgd_r30_s42_p3_streamingllm_scored.jsonl | Data | 132.0 KB | — |
| results/lme_ku/lme_ku_s42_r10_n72_sdpa_aggregate.json | Data | 1.4 KB | — |
| results/lme_ku/lme_ku_s42_r10_n72_sdpa_reader_shard0.jsonl | Data | 244.1 KB | — |
| results/lme_ku/lme_ku_s42_r10_n72_sdpa_scored.jsonl | Data | 195.8 KB | — |
| results/lme_ku/lme_ku_s42_r30_n72_sdpa_aggregate.json | Data | 1.4 KB | — |
| results/lme_ku/lme_ku_s42_r30_n72_sdpa_reader_shard0.jsonl | Data | 245.4 KB | — |
| results/lme_ku/lme_ku_s42_r30_n72_sdpa_scored.jsonl | Data | 195.2 KB | — |
| results/lme_ku/lme_ku_s43_r10_n72_sdpa_aggregate.json | Data | 1.4 KB | — |
| results/lme_ku/lme_ku_s43_r10_n72_sdpa_reader_shard0.jsonl | Data | 244.1 KB | — |
| results/lme_ku/lme_ku_s43_r10_n72_sdpa_scored.jsonl | Data | 195.8 KB | — |
| results/lme_ku/lme_ku_s43_r30_n72_sdpa_aggregate.json | Data | 1.4 KB | — |
| results/lme_ku/lme_ku_s43_r30_n72_sdpa_reader_shard0.jsonl | Data | 245.4 KB | — |
| results/lme_ku/lme_ku_s43_r30_n72_sdpa_scored.jsonl | Data | 195.2 KB | — |
| results/main/chunkkv_mw_r30_p1_aggregate.json | Data | 182 B | — |
| results/main/chunkkv_mw_r30_p3_aggregate.json | Data | 309 B | — |
| results/main/chunkkv_mw_r30_p3_scored.jsonl | Data | 164.4 KB | — |
| results/main/chunkkv_sgd_r30_p1_aggregate.json | Data | 182 B | — |
| results/main/chunkkv_sgd_r30_p3_aggregate.json | Data | 304 B | — |
| results/main/chunkkv_sgd_r30_p3_scored.jsonl | Data | 130.3 KB | — |
| results/main/mw_r10_s42_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r10_s42_p1_tokenf1_audit_aggregate.json | Data | 1.8 KB | — |
| results/main/mw_r10_s42_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/mw_r10_s42_p3_scored.jsonl | Data | 965.7 KB | — |
| results/main/mw_r10_s43_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r10_s43_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/mw_r10_s43_p3_scored.jsonl | Data | 1.0 MB | — |
| results/main/mw_r10_s44_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r10_s44_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/mw_r10_s44_p3_scored.jsonl | Data | 1.0 MB | — |
| results/main/mw_r30_s42_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r30_s42_p1_tokenf1_audit_aggregate.json | Data | 1.8 KB | — |
| results/main/mw_r30_s42_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/mw_r30_s42_p3_scored.jsonl | Data | 1.2 MB | — |
| results/main/mw_r30_s43_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r30_s43_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/mw_r30_s43_p3_scored.jsonl | Data | 1.0 MB | — |
| results/main/mw_r30_s44_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r30_s44_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/mw_r30_s44_p3_scored.jsonl | Data | 1.0 MB | — |
| results/main/mw_r50_s42_aggregate.json | Data | 1.3 KB | — |
| results/main/mw_r50_s42_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r50_s42_p3_scored.jsonl | Data | 523.5 KB | — |
| results/main/mw_r50_s43_aggregate.json | Data | 858 B | — |
| results/main/mw_r50_s43_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r50_s43_p3_scored.jsonl | Data | 545.6 KB | — |
| results/main/mw_r50_s44_aggregate.json | Data | 1.3 KB | — |
| results/main/mw_r50_s44_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r50_s44_p3_scored.jsonl | Data | 547.6 KB | — |
| results/main/mw_r70_s42_aggregate.json | Data | 1.3 KB | — |
| results/main/mw_r70_s42_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r70_s42_p3_scored.jsonl | Data | 523.8 KB | — |
| results/main/mw_r70_s43_aggregate.json | Data | 854 B | — |
| results/main/mw_r70_s43_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r70_s43_p3_scored.jsonl | Data | 545.7 KB | — |
| results/main/mw_r70_s44_aggregate.json | Data | 1.3 KB | — |
| results/main/mw_r70_s44_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/mw_r70_s44_p3_scored.jsonl | Data | 547.9 KB | — |
| results/main/p1_token_f1_audit.json | Data | 5.7 KB | — |
| results/main/seed_summary.json | Data | 101.0 KB | — |
| results/main/seed_summary_wider.json | Data | 71.3 KB | — |
| results/main/sgd_r10_s42_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r10_s42_p1_tokenf1_audit_aggregate.json | Data | 1.8 KB | — |
| results/main/sgd_r10_s42_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/sgd_r10_s42_p3_scored.jsonl | Data | 759.7 KB | — |
| results/main/sgd_r10_s43_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r10_s43_p3_aggregate.json | Data | 2.1 KB | — |
| results/main/sgd_r10_s43_p3_scored.jsonl | Data | 738.1 KB | — |
| results/main/sgd_r10_s44_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r10_s44_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/sgd_r10_s44_p3_scored.jsonl | Data | 778.5 KB | — |
| results/main/sgd_r30_s42_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r30_s42_p1_tokenf1_audit_aggregate.json | Data | 1.8 KB | — |
| results/main/sgd_r30_s42_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/sgd_r30_s42_p3_scored.jsonl | Data | 946.9 KB | — |
| results/main/sgd_r30_s43_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r30_s43_p3_aggregate.json | Data | 2.1 KB | — |
| results/main/sgd_r30_s43_p3_scored.jsonl | Data | 739.0 KB | — |
| results/main/sgd_r30_s44_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r30_s44_p3_aggregate.json | Data | 2.2 KB | — |
| results/main/sgd_r30_s44_p3_scored.jsonl | Data | 779.2 KB | — |
| results/main/sgd_r50_s42_aggregate.json | Data | 1.3 KB | — |
| results/main/sgd_r50_s42_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r50_s42_p3_scored.jsonl | Data | 411.8 KB | — |
| results/main/sgd_r50_s43_aggregate.json | Data | 848 B | — |
| results/main/sgd_r50_s43_p3_aggregate.json | Data | 1.3 KB | — |
| results/main/sgd_r50_s43_p3_scored.jsonl | Data | 399.9 KB | — |
| results/main/sgd_r50_s44_aggregate.json | Data | 1.3 KB | — |
| results/main/sgd_r50_s44_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r50_s44_p3_scored.jsonl | Data | 421.7 KB | — |
| results/main/sgd_r70_s42_aggregate.json | Data | 1.3 KB | — |
| results/main/sgd_r70_s42_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r70_s42_p3_scored.jsonl | Data | 410.9 KB | — |
| results/main/sgd_r70_s43_aggregate.json | Data | 843 B | — |
| results/main/sgd_r70_s43_p3_aggregate.json | Data | 1.3 KB | — |
| results/main/sgd_r70_s43_p3_scored.jsonl | Data | 399.4 KB | — |
| results/main/sgd_r70_s44_aggregate.json | Data | 1.3 KB | — |
| results/main/sgd_r70_s44_p3_aggregate.json | Data | 1.4 KB | — |
| results/main/sgd_r70_s44_p3_scored.jsonl | Data | 421.3 KB | — |
| results/p1_saved/mw_r10_s42_scored.jsonl | Data | 483.0 KB | — |
| results/p1_saved/mw_r10_s43_scored.jsonl | Data | 479.5 KB | — |
| results/p1_saved/mw_r10_s44_scored.jsonl | Data | 484.5 KB | — |
| results/p1_saved/mw_r30_s42_scored.jsonl | Data | 482.1 KB | — |
| results/p1_saved/mw_r30_s43_scored.jsonl | Data | 479.8 KB | — |
| results/p1_saved/mw_r30_s44_scored.jsonl | Data | 484.2 KB | — |
| results/p1_saved/provenance.json | Data | 1.9 KB | — |
| results/p1_saved/sgd_r10_s42_scored.jsonl | Data | 428.6 KB | — |
| results/p1_saved/sgd_r10_s43_scored.jsonl | Data | 435.1 KB | — |
| results/p1_saved/sgd_r10_s44_scored.jsonl | Data | 430.3 KB | — |
| results/p1_saved/sgd_r30_s42_scored.jsonl | Data | 429.0 KB | — |
| results/p1_saved/sgd_r30_s43_scored.jsonl | Data | 435.7 KB | — |
| results/p1_saved/sgd_r30_s44_scored.jsonl | Data | 430.8 KB | — |
| results/p3_segment_sensitivity/changed_answers.jsonl | Data | 628.5 KB | — |
| results/p3_segment_sensitivity/per_answer.jsonl | Data | 12.0 MB | e79a208279f3 |
| results/p3_segment_sensitivity/summary.json | Data | 137.8 KB | — |
| results/risawoz/risawoz_s42_r10_n200_aggregate.json | Data | 1.7 KB | — |
| results/risawoz/risawoz_s42_r30_n200_aggregate.json | Data | 1.7 KB | — |
| results/seed_robustness/aggregates/lme_ku_p2.json | Data | 2.5 KB | — |
| results/seed_robustness/aggregates/risawoz_p1.json | Data | 2.8 KB | — |
| results/seed_robustness/aggregates/risawoz_p2.json | Data | 2.4 KB | — |
| results/seed_robustness/experiment.json | Data | 2.7 KB | — |
| results/seed_robustness/manifest.json | Data | 2.9 KB | — |
| results/seed_robustness/reader/lme_ku_all72_r30_llama.jsonl | Data | 402.6 KB | — |
| results/seed_robustness/reader/risawoz_all207_r30_llama.jsonl | Data | 1.9 MB | — |
| results/seed_robustness/scored/lme_ku_p2.jsonl | Data | 351.6 KB | — |
| results/seed_robustness/scored/risawoz_p1.jsonl | Data | 981.2 KB | — |
| results/seed_robustness/scored/risawoz_p2.jsonl | Data | 1.0 MB | — |
| results/seed_robustness/stats/summary.json | Data | 19.7 KB | — |
| results/semantic_evaluation/ae_development/execution_summary.json | Data | 358 B | — |
| results/semantic_evaluation/ae_development/judgments.jsonl | Data | 53.0 KB | — |
| results/semantic_evaluation/ae_development/model_record.json | Data | 719 B | — |
| results/semantic_evaluation/ae_development/raw_responses.jsonl | Data | 1.4 MB | — |
| results/semantic_evaluation/ae_development_metrics.json | Data | 6.8 KB | — |
| results/semantic_evaluation/ae_test/execution_summary.json | Data | 391 B | — |
| results/semantic_evaluation/ae_test/judgments.jsonl | Data | 159.7 KB | — |
| results/semantic_evaluation/ae_test/model_record.json | Data | 719 B | — |
| results/semantic_evaluation/ae_test/raw_responses.jsonl | Data | 4.1 MB | — |
| results/semantic_evaluation/ae_test_metrics.json | Data | 6.8 KB | — |
| results/semantic_evaluation/p1/execution_summary.json | Data | 377 B | — |
| results/semantic_evaluation/p1/judgments.jsonl | Data | 965.2 KB | — |
| results/semantic_evaluation/p1/model_record.json | Data | 719 B | — |
| results/semantic_evaluation/p1/raw_responses.jsonl | Data | 21.4 MB | 754459097f00 |
| results/semantic_evaluation/p1_action_domain/execution_summary.json | Data | 463 B | — |
| results/semantic_evaluation/p1_action_domain/judgments.jsonl | Data | 731.0 KB | — |
| results/semantic_evaluation/p1_action_domain/model_record.json | Data | 719 B | — |
| results/semantic_evaluation/p1_action_domain/raw_responses.jsonl | Data | 20.6 MB | e19b07f1de98 |
| results/semantic_evaluation/p1_report/rank_and_disagreement_analysis.json | Data | 8.0 KB | — |
| results/semantic_evaluation/p1_summary.json | Data | 34.0 KB | — |
| results/semantic_evaluation/rubric_controls/conformance.json | Data | 2.0 KB | — |
| results/semantic_evaluation/rubric_controls/execution_summary.json | Data | 346 B | — |
| results/semantic_evaluation/rubric_controls/judgments.jsonl | Data | 3.4 KB | — |
| results/semantic_evaluation/rubric_controls/model_record.json | Data | 719 B | — |
| results/semantic_evaluation/rubric_controls/raw_responses.jsonl | Data | 113.0 KB | — |
| LICENSE | Documentation | 1.8 KB | — |
| README.md | Documentation | 8.4 KB | — |
| RESULTS_MAP.md | Documentation | 4.0 KB | — |
| SEMANTIC_EVALUATION.md | Documentation | 7.3 KB | — |
| results/p3_segment_sensitivity/REPORT.md | Documentation | 3.0 KB | — |
| results/seed_robustness/stats/summary.md | Documentation | 1.4 KB | — |
| results/semantic_evaluation/p1_report/REPORT.md | Documentation | 2.2 KB | — |
| scripts/fmts_refpool/README.md | Documentation | 2.6 KB | — |
| run_all.sh | Other | 550 B | — |
| scripts/audit/anchor_filter_audit.py | Other | 5.1 KB | — |
| scripts/audit/rank_inconsistency_final_slot.py | Other | 4.7 KB | — |
| scripts/audit/recompute_p1_audit.py | Other | 2.9 KB | — |
| scripts/audit/validate_fmts_support_ablation.py | Other | 10.8 KB | — |
| scripts/audit/validate_saved_readers.py | Other | 4.7 KB | — |
| scripts/build_aggregates_flat.py | Other | 10.6 KB | — |
| scripts/builders/build_deletions.py | Other | 2.2 KB | — |
| scripts/builders/build_lme_probes.py | Other | 7.7 KB | — |
| scripts/builders/build_lme_transitions.py | Other | 9.9 KB | — |
| scripts/builders/build_probes.py | Other | 4.6 KB | — |
| scripts/builders/build_risawoz_probes.py | Other | 13.8 KB | — |
| scripts/builders/construction.py | Other | 6.7 KB | — |
| scripts/check_artifact.py | Other | 1.8 KB | — |
| scripts/fmts_refpool/core/boundary_detector.py | Other | 6.2 KB | — |
| scripts/fmts_refpool/core/compression_methods.py | Other | 6.3 KB | — |
| scripts/fmts_refpool/core/evaluation.py | Other | 6.3 KB | — |
| scripts/fmts_refpool/core/turn_types.py | Other | 154 B | — |
| scripts/fmts_refpool/run_answers.py | Other | 3.2 KB | — |
| scripts/fmts_refpool/run_compression.py | Other | 10.9 KB | — |
| scripts/normalize.py | Other | 5.8 KB | — |
| scripts/reader/chunkkv_vendored/__init__.py | Other | 1.0 KB | — |
| scripts/reader/chunkkv_vendored/base_press.py | Other | 4.1 KB | — |
| scripts/reader/chunkkv_vendored/chunkkv_press.py | Other | 4.8 KB | — |
| scripts/reader/chunkkv_vendored/scorer_press.py | Other | 1.8 KB | — |
| scripts/reader/chunkkv_vendored/snapkv_press.py | Other | 3.3 KB | — |
| scripts/reader/chunkkv_vendored/utils.py | Other | 3.3 KB | — |
| scripts/reader/reader_worker_chunkkv.py | Other | 15.1 KB | — |
| scripts/reader/reader_worker_kvpress_extra.py | Other | 16.5 KB | — |
| scripts/reader/reader_worker_local.py | Other | 8.4 KB | — |
| scripts/scorer/scorer_lme_ku.py | Other | 10.0 KB | — |
| scripts/scorer/scorer_p1.py | Other | 11.1 KB | — |
| scripts/scorer/scorer_p1_late.py | Other | 12.9 KB | — |
| scripts/scorer/scorer_p1_zh.py | Other | 6.1 KB | — |
| scripts/scorer/scorer_p3.py | Other | 8.2 KB | — |
| scripts/seed_robustness/bootstrap_stats.py | Other | 10.6 KB | — |
| scripts/seed_robustness/build_seed_robustness.py | Other | 7.0 KB | — |
| scripts/seed_robustness/make_manifest.py | Other | 2.8 KB | — |
| scripts/seed_robustness/validate_seed_robustness.py | Other | 4.3 KB | — |
| scripts/semantic_evaluation/aggregate_ae_judge.py | Other | 4.3 KB | — |
| scripts/semantic_evaluation/aggregate_p1_semantic.py | Other | 5.9 KB | — |
| scripts/semantic_evaluation/audit_p3_segments.py | Other | 11.0 KB | — |
| scripts/semantic_evaluation/check_core_rubric.py | Other | 2.5 KB | — |
| scripts/semantic_evaluation/prepare_ae_validation.py | Other | 6.4 KB | — |
| scripts/semantic_evaluation/prepare_p1_semantic.py | Other | 3.0 KB | — |
| scripts/semantic_evaluation/recover_request_ids.py | Other | 2.1 KB | — |
| scripts/semantic_evaluation/report_p1_semantic.py | Other | 5.4 KB | — |
| scripts/semantic_evaluation/run_core_goal_judge.py | Other | 670 B | — |
| scripts/semantic_evaluation/run_semantic_judge.py | Other | 8.4 KB | — |
| scripts/semantic_evaluation/validate_evaluation_outputs.py | Other | 3.2 KB | — |
| scripts/tests/test_construction.py | Other | 1.6 KB | — |
| scripts/tests/test_corpus_construction.py | Other | 1.3 KB | — |
| scripts/tests/test_fmts.py | Other | 1.7 KB | — |
| scripts/validate_corpus_scorers.py | Other | 2.9 KB | — |
| .gitattributes | Repository | 3.2 KB | — |
License and Download
- License
- other
- Access
- No access gate
Released by Papersubmission through its official repository on Hugging Face.