{ "counts": { "train": 575, "valid": 124, "test": 124, "train_vocab": 39, "test_vocab": 35, "unseen_test_tools": [ "mcp__cowork__request_cowork_directory" ] }, "top_train_tools": [ [ "Bash", 482 ], [ "Read", 387 ], [ "Edit", 218 ], [ "Write", 97 ], [ "StructuredOutput", 87 ], [ "mcp__ccd_session__mark_chapter", 85 ], [ "run_shell_command", 56 ], [ "replace", 33 ], [ "ToolSearch", 31 ], [ "read_file", 21 ], [ "grep_search", 18 ], [ "Agent", 17 ], [ "TaskUpdate", 16 ], [ "Grep", 12 ], [ "Monitor", 12 ], [ "TaskStop", 10 ], [ "ScheduleWakeup", 9 ], [ "mcp__ccd_session__spawn_task", 8 ], [ "mcp__playwright__browser_navigate", 6 ], [ "mcp__playwright__browser_take_screenshot", 6 ] ], "top_test_tools": [ [ "Bash", 93 ], [ "Read", 75 ], [ "Edit", 55 ], [ "StructuredOutput", 23 ], [ "mcp__ccd_session__mark_chapter", 15 ], [ "Write", 14 ], [ "run_shell_command", 13 ], [ "read_file", 9 ], [ "Agent", 7 ], [ "ToolSearch", 7 ], [ "replace", 5 ], [ "mcp__playwright__browser_navigate", 5 ], [ "grep_search", 4 ], [ "mcp__playwright__browser_take_screenshot", 4 ], [ "Monitor", 4 ], [ "mcp__playwright__browser_resize", 3 ], [ "mcp__playwright__browser_evaluate", 3 ], [ "list_directory", 3 ], [ "TaskUpdate", 3 ], [ "TaskStop", 3 ] ], "floor_by_k": [ { "k": 1, "tools": [ "Bash" ], "jaccard": 0.31837557603686634, "exact": 0.12096774193548387, "precision": 0.75, "recall": 0.31837557603686634, "top1": 0.75 }, { "k": 2, "tools": [ "Bash", "Read" ], "jaccard": 0.43137480798771116, "exact": 0.08870967741935484, "precision": 0.6774193548387096, "recall": 0.5031490015360983, "top1": 0.75 }, { "k": 3, "tools": [ "Bash", "Read", "Edit" ], "jaccard": 0.4740975422427035, "exact": 0.11290322580645161, "precision": 0.5994623655913978, "recall": 0.6307987711213517, "top1": 0.75 }, { "k": 4, "tools": [ "Bash", "Read", "Edit", "Write" ], "jaccard": 0.4057027649769585, "exact": 0.016129032258064516, "precision": 0.4778225806451613, "recall": 0.6584965437788018, "top1": 0.75 }, { "k": 5, "tools": [ "Bash", "Read", "Edit", "Write", "StructuredOutput" ], "jaccard": 0.3831989247311828, "exact": 0.0, "precision": 0.41935483870967744, "recall": 0.714544930875576, "top1": 0.75 }, { "k": 6, "tools": [ "Bash", "Read", "Edit", "Write", "StructuredOutput", "mcp__ccd_session__mark_chapter" ], "jaccard": 0.3497064539403249, "exact": 0.0, "precision": 0.3696236559139785, "recall": 0.7408986175115208, "top1": 0.75 }, { "k": 7, "tools": [ "Bash", "Read", "Edit", "Write", "StructuredOutput", "mcp__ccd_session__mark_chapter", "run_shell_command" ], "jaccard": 0.3145152562491272, "exact": 0.0, "precision": 0.3317972350230414, "recall": 0.8076996927803379, "top1": 0.75 }, { "k": 8, "tools": [ "Bash", "Read", "Edit", "Write", "StructuredOutput", "mcp__ccd_session__mark_chapter", "run_shell_command", "replace" ], "jaccard": 0.2813046720304785, "exact": 0.0, "precision": 0.29536290322580644, "recall": 0.8220814132104455, "top1": 0.75 } ], "summary": { "diffusion_zero_shot": { "jaccard": [ 0.07278545826932924, 0.044809267793138756, 0.10379544290834614 ], "top1": 0.0967741935483871, "exact": 0.008064516129032258, "precision": 0.12916666666666665, "recall": 0.09401881720430108, "pred_n": 1.7741935483870968, "missing": 2.685483870967742, "extra": 1.4919354838709677 }, "diffusion_lora": { "jaccard": [ 0.4472606246799795, 0.3933147721454173, 0.5016513056835638 ], "top1": 0.75, "exact": 0.10483870967741936, "precision": 0.5658602150537635, "recall": 0.6065956221198157, "pred_n": 2.879032258064516, "missing": 1.2661290322580645, "extra": 1.1774193548387097 }, "ar_zero_shot": { "jaccard": [ 0.07011919532080822, 0.04485037811650715, 0.0987386269644334 ], "top1": 0.11290322580645161, "exact": 0.008064516129032258, "precision": 0.12575475599669148, "recall": 0.11344086021505376, "pred_n": 2.379032258064516, "missing": 2.6693548387096775, "extra": 2.0806451612903225 }, "ar_lora": { "jaccard": [ 0.4925115207373272, 0.4320468509984639, 0.5533218125960062 ], "top1": 0.782258064516129, "exact": 0.22580645161290322, "precision": 0.7553763440860215, "recall": 0.5109927035330262, "pred_n": 1.75, "missing": 1.5725806451612903, "extra": 0.3548387096774194 }, "qwen35_zero_shot": { "jaccard": [ 0.1381896637945025, 0.10233294930875576, 0.17591047792660694 ], "top1": 0.25, "exact": 0.008064516129032258, "precision": 0.21700372607630672, "recall": 0.21192396313364056, "pred_n": 2.4274193548387095, "missing": 2.2903225806451615, "extra": 1.75 }, "qwen27_zero_shot": { "jaccard": [ 0.19682306940371458, 0.15791788856304986, 0.23646662477307637 ], "top1": 0.33064516129032256, "exact": 0.008064516129032258, "precision": 0.322599846390169, "recall": 0.279531490015361, "pred_n": 2.282258064516129, "missing": 2.1048387096774195, "extra": 1.4193548387096775 }, "diffusion_aw005": { "jaccard": [ 0.4254032258064516, 0.37307987711213514, 0.4777073732718894 ], "top1": 0.7580645161290323, "exact": 0.08870967741935484, "precision": 0.5090693804403482, "recall": 0.6351478494623657, "pred_n": 3.467741935483871, "missing": 1.1612903225806452, "extra": 1.6612903225806452 }, "diffusion_aw025": { "jaccard": [ 0.43988491347362313, 0.38916963412931155, 0.4930299539170507 ], "top1": 0.75, "exact": 0.08064516129032258, "precision": 0.542388928276025, "recall": 0.6325940860215054, "pred_n": 3.411290322580645, "missing": 1.185483870967742, "extra": 1.6290322580645162 }, "diffusion_r32": { "jaccard": [ 0.44411802355350743, 0.3898361495135689, 0.49879032258064515 ], "top1": 0.75, "exact": 0.12096774193548387, "precision": 0.668010752688172, "recall": 0.5104646697388633, "pred_n": 2.064516129032258, "missing": 1.532258064516129, "extra": 0.6290322580645161 }, "diffusion_expanded": { "jaccard": [ 0.4335445468509984, 0.3826612903225806, 0.48529185867895547 ], "top1": 0.7580645161290323, "exact": 0.08064516129032258, "precision": 0.6024193548387097, "recall": 0.5542338709677419, "pred_n": 2.475806451612903, "missing": 1.435483870967742, "extra": 0.9435483870967742 }, "floor_top3": { "jaccard": [ 0.4740975422427035, 0.4201996927803379, 0.5277265745007681 ], "top1": 0.75, "exact": 0.11290322580645161, "precision": 0.5994623655913978, "recall": 0.6307987711213517, "pred_n": 3.0, "missing": 1.1693548387096775, "extra": 1.2016129032258065 }, "tfidf_1nn": { "jaccard": [ 0.500444828469022, 0.4454237071172555, 0.5572644649257552 ], "top1": 0.7661290322580645, "exact": 0.1532258064516129, "precision": 0.6579781105990783, "recall": 0.639276113671275, "pred_n": 2.814516129032258, "missing": 1.1451612903225807, "extra": 0.9919354838709677 }, "tfidf_1nn_top3vote": { "jaccard": [ 0.48733678955453147, 0.4321716589861751, 0.5448636712749616 ], "top1": 0.7661290322580645, "exact": 0.1693548387096774, "precision": 0.6760752688172043, "recall": 0.59176267281106, "pred_n": 2.2903225806451615, "missing": 1.3548387096774193, "extra": 0.6774193548387096 }, "tfidf_3nn": { "jaccard": [ 0.5050179211469534, 0.45102406554019453, 0.5590437788018433 ], "top1": 0.7903225806451613, "exact": 0.1693548387096774, "precision": 0.6533602150537634, "recall": 0.6464285714285715, "pred_n": 2.7419354838709675, "missing": 1.1290322580645162, "extra": 0.9032258064516129 }, "tfidf_3nn_top3vote": { "jaccard": [ 0.510515873015873, 0.45830773169482847, 0.5629704301075269 ], "top1": 0.7903225806451613, "exact": 0.16129032258064516, "precision": 0.6370967741935484, "recall": 0.6839093701996928, "pred_n": 2.8548387096774195, "missing": 1.1290322580645162, "extra": 1.0161290322580645 }, "tfidf_5nn": { "jaccard": [ 0.500502432155658, 0.4513536866359447, 0.5495967741935485 ], "top1": 0.7903225806451613, "exact": 0.12096774193548387, "precision": 0.637768817204301, "recall": 0.6762576804915514, "pred_n": 2.7983870967741935, "missing": 1.1048387096774193, "extra": 0.9354838709677419 }, "tfidf_5nn_top3vote": { "jaccard": [ 0.5122855862775217, 0.45913978494623653, 0.5648841525857655 ], "top1": 0.7903225806451613, "exact": 0.1774193548387097, "precision": 0.6303763440860215, "recall": 0.6996351766513057, "pred_n": 2.935483870967742, "missing": 1.096774193548387, "extra": 1.064516129032258 }, "train_labelset_oracle": { "jaccard": [ 0.9444124423963134, 0.9213997695852535, 0.9657834101382489 ], "top1": 0.9354838709677419, "exact": 0.8145161290322581, "precision": 0.9692492319508448, "recall": 0.9717069892473118, "pred_n": 3.0, "missing": 0.10483870967741936, "extra": 0.13709677419354838 }, "data_ceiling": { "jaccard": [ 0.9959677419354839, 0.9879032258064516, 1.0 ], "top1": 1.0, "exact": 0.9919354838709677, "precision": 1.0, "recall": 0.9959677419354839, "pred_n": 2.9596774193548385, "missing": 0.008064516129032258, "extra": 0.0 } }, "paired_vs_floor": { "diffusion_zero_shot": [ -0.4013120839733743, -0.4626536098310292, -0.338773681515617 ], "diffusion_lora": [ -0.02683691756272401, -0.06750192012288786, 0.012679211469534053 ], "ar_zero_shot": [ -0.4039783469218953, -0.46681436842727164, -0.33903609831029186 ], "ar_lora": [ 0.01841397849462366, -0.03867127496159754, 0.07795698924731184 ], "qwen35_zero_shot": [ -0.33590787844820097, -0.39298195084485404, -0.27843161814936007 ], "qwen27_zero_shot": [ -0.277274472838989, -0.3321507121910348, -0.2189315388912163 ], "diffusion_aw005": [ -0.04869431643625192, -0.0894489247311828, -0.008787762416794671 ], "diffusion_aw025": [ -0.034212628769080385, -0.06047467317628608, -0.007286866359447002 ], "diffusion_r32": [ -0.029979518689196107, -0.08051075268817205, 0.021873399897593446 ], "diffusion_expanded": [ -0.040552995391705066, -0.07831221198156682, -0.0032258064516129024 ], "tfidf_1nn": [ 0.026347286226318482, -0.03129800307219662, 0.08436059907834101 ], "tfidf_1nn_top3vote": [ 0.013239247311827957, -0.04418202764976958, 0.07288786482334869 ], "tfidf_3nn": [ 0.030920378904249875, -0.0268721198156682, 0.0878200204813108 ], "tfidf_3nn_top3vote": [ 0.036418330773169486, -0.010615079365079364, 0.08473502304147465 ], "tfidf_5nn": [ 0.02640488991295443, -0.026379288274449565, 0.07841461853558628 ], "tfidf_5nn_top3vote": [ 0.03818804403481822, -0.005081925243215566, 0.08151881720430107 ], "train_labelset_oracle": [ 0.4703149001536099, 0.4196524577572965, 0.5210253456221199 ], "data_ceiling": [ 0.5218701996927804, 0.468394777265745, 0.5749135944700461 ] }, "strata_counts": { "freq": { "head_only": 44, "one_tail": 48, "multi_tail": 32 }, "card": { "1-2": 50, "3": 35, "4+": 39 }, "length": { "short": 42, "medium": 41, "long": 41 } } }