Hugging Face's logo

Datasets:
hermionai
/
hermion-qmsum-interaction-dynamics

Tasks:
Text Classification
Other
Modalities:
Text
Languages:
English
Size:
n<1K
Tags:
interaction-dynamics
team-intelligence
organizational-behavior
meeting-intelligence
group-intelligence
physics-based
License:
Dataset card Data Studio Files
xet
 Community
Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code:   DatasetGenerationCastError
Exception:    DatasetGenerationCastError
Message:      An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 1 new columns ({'anchor'}) and 1 missing columns ({'session_turns'}).

This happened while the csv dataset builder was generating data using

hf://datasets/hermionai/hermion-qmsum-interaction-dynamics/summary_study3_longitudinal.csv (at revision 7921c36e6496206c04d0bcff7b46bd2a724ca042), ['hf://datasets/hermionai/hermion-qmsum-interaction-dynamics@7921c36e6496206c04d0bcff7b46bd2a724ca042/summary_study1_longitudinal.csv', 'hf://datasets/hermionai/hermion-qmsum-interaction-dynamics@7921c36e6496206c04d0bcff7b46bd2a724ca042/summary_study2_longitudinal.csv', 'hf://datasets/hermionai/hermion-qmsum-interaction-dynamics@7921c36e6496206c04d0bcff7b46bd2a724ca042/summary_study3_longitudinal.csv']

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1837, in _prepare_split_single
                  writer.write_table(table)
                  ~~~~~~~~~~~~~~~~~~^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 765, in write_table
                  self._write_table(pa_table, writer_batch_size=writer_batch_size)
                  ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 773, in _write_table
                  pa_table = table_cast(pa_table, self._schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2369, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              study: string
              run_label: string
              meeting_id: string
              phase: string
              total_speakers: int64
              assigned_speakers: int64
              ignored_speakers: int64
              group_a: string
              group_b: string
              raw_turns: int64
              clean_turns: int64
              prepared_messages: int64
              signal_count: int64
              progression_pct: int64
              peak_pct: int64
              momentum: string
              interest_level: string
              topology_mode: string
              topology_label: string
              recovery_capacity: string
              primary_resistance: string
              dominant_component: string
              temporal_share: int64
              structural_share: double
              asymmetric_share: double
              ghosting_risk: double
              rejection_risk: double
              stall_risk: double
              self_positive_ratio: double
              other_positive_ratio: double
              self_withdrawing: bool
              other_withdrawing: bool
              self_signal_share: double
              next_move: string
              other_next_move: string
              mismatch_detected: bool
              mismatch_desc: string
              processed_at: string
              anchor: string
              session_count: int64
              -- schema metadata --
              pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 5281
              to
              {'study': Value('string'), 'run_label': Value('string'), 'meeting_id': Value('string'), 'phase': Value('string'), 'total_speakers': Value('int64'), 'assigned_speakers': Value('int64'), 'ignored_speakers': Value('int64'), 'group_a': Value('string'), 'group_b': Value('string'), 'raw_turns': Value('int64'), 'clean_turns': Value('int64'), 'prepared_messages': Value('int64'), 'signal_count': Value('int64'), 'progression_pct': Value('int64'), 'peak_pct': Value('int64'), 'momentum': Value('string'), 'interest_level': Value('string'), 'topology_mode': Value('string'), 'topology_label': Value('string'), 'recovery_capacity': Value('string'), 'primary_resistance': Value('string'), 'dominant_component': Value('string'), 'temporal_share': Value('int64'), 'structural_share': Value('float64'), 'asymmetric_share': Value('float64'), 'ghosting_risk': Value('float64'), 'rejection_risk': Value('float64'), 'stall_risk': Value('float64'), 'self_positive_ratio': Value('float64'), 'other_positive_ratio': Value('float64'), 'self_withdrawing': Value('bool'), 'other_withdrawing': Value('bool'), 'self_signal_share': Value('float64'), 'next_move': Value('string'), 'other_next_move': Value('string'), 'mismatch_detected': Value('bool'), 'mismatch_desc': Value('string'), 'processed_at': Value('string'), 'session_count': Value('int64'), 'session_turns': Value('string')}
              because column names don't match
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1683, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1839, in _prepare_split_single
                  raise DatasetGenerationCastError.from_cast_error(
                  ...<4 lines>...
                  )
              datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
              
              All the data files must have the same columns, but at some point there are 1 new columns ({'anchor'}) and 1 missing columns ({'session_turns'}).
              
              This happened while the csv dataset builder was generating data using
              
              hf://datasets/hermionai/hermion-qmsum-interaction-dynamics/summary_study3_longitudinal.csv (at revision 7921c36e6496206c04d0bcff7b46bd2a724ca042), ['hf://datasets/hermionai/hermion-qmsum-interaction-dynamics@7921c36e6496206c04d0bcff7b46bd2a724ca042/summary_study1_longitudinal.csv', 'hf://datasets/hermionai/hermion-qmsum-interaction-dynamics@7921c36e6496206c04d0bcff7b46bd2a724ca042/summary_study2_longitudinal.csv', 'hf://datasets/hermionai/hermion-qmsum-interaction-dynamics@7921c36e6496206c04d0bcff7b46bd2a724ca042/summary_study3_longitudinal.csv']
              
              Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

study
string
run_label
string
meeting_id
string
phase
string
total_speakers
int64
assigned_speakers
int64
ignored_speakers
int64
group_a
string
group_b
string
raw_turns
int64
clean_turns
int64
prepared_messages
int64
signal_count
int64
progression_pct
int64
peak_pct
int64
momentum
string
interest_level
string
topology_mode
string
topology_label
string
recovery_capacity
string
primary_resistance
string
dominant_component
string
temporal_share
int64
structural_share
float64
asymmetric_share
float64
ghosting_risk
float64
rejection_risk
float64
stall_risk
float64
self_positive_ratio
float64
other_positive_ratio
float64
self_withdrawing
bool
other_withdrawing
bool
self_signal_share
float64
next_move
string
other_next_move
string
mismatch_detected
bool
mismatch_desc
string
processed_at
string
session_count
int64
session_turns
string
study1_longitudinal
Management_vs_Design
ES2002
longitudinal
4
4
0
Design
Management
2,498
2,340
2,340
2,372
7
11
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.608
0.392
0.804
0.413
0.88
0.077
0.143
false
false
0.65
reengage
reciprocate
false
null
2026-07-23T00:30:51.458299
4
[256, 589, 603, 892]
study1_longitudinal
Management_vs_Design
ES2003
longitudinal
4
4
0
Design
Management
1,281
1,204
1,204
1,221
5
17
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.817
0.183
0.844
0.44
0.9
0.2
0
false
true
0.75
pull_back
continue_withdrawal
true
Before declines, Management leans toward withdrawal while Design leans toward withdrawal. Before recoveries, the pattern shifts to Management leaning bonding and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:31:03.932180
4
[111, 307, 306, 480]
study1_longitudinal
Management_vs_Design
ES2004
longitudinal
4
4
0
Design
Management
2,208
2,069
2,069
2,091
5
35
declining
medium
phase_plus_quantum
declining
medium
structural friction
r_structural
0
0.73
0.27
0.842
0.449
0.9
0.091
0.222
false
false
0.55
reengage
reciprocate
true
Before declines, Management leans toward bonding while Design leans toward withdrawal. Before recoveries, the pattern shifts to Management leaning withdrawal and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:31:17.793187
4
[291, 502, 578, 698]
study1_longitudinal
Management_vs_Design
ES2005
longitudinal
4
4
0
Design
Management
2,160
1,901
1,901
1,924
2
2
gliding
low
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.938
0.524
0.96
0.177
0.154
false
false
0.507
reengage
reciprocate
false
null
2026-07-23T00:31:30.685359
4
[93, 569, 750, 489]
study1_longitudinal
Management_vs_Design
ES2006
longitudinal
4
4
0
Design
Management
2,432
2,182
2,182
2,212
8
13
declining
medium
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.876
0.124
0.8
0.394
0.88
0.385
0.571
true
false
0.65
increase_investment
wait_and_test
true
Before declines, Management leans toward deal progression while Design leans toward withdrawal. Before recoveries, the pattern shifts to Management leaning withdrawal and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:31:42.191487
4
[243, 506, 623, 810]
study1_longitudinal
Management_vs_Design
ES2007
longitudinal
4
4
0
Design
Management
1,708
1,501
1,501
1,522
7
14
declining
medium
phase_plus_quantum
declining
medium
structural friction
r_structural
0
0.892
0.108
0.803
0.393
0.88
0.214
0.333
false
true
0.7
change_tone
continue_withdrawal
false
null
2026-07-23T00:31:52.369936
4
[284, 334, 514, 369]
study1_longitudinal
Management_vs_Design
ES2008
longitudinal
4
4
0
Design
Management
2,027
1,832
1,832
1,868
8
22
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.749
0.251
0.776
0.383
0.86
0.308
0.286
false
false
0.65
reengage
reciprocate
false
null
2026-07-23T00:32:04.109689
4
[162, 437, 439, 794]
study1_longitudinal
Management_vs_Design
ES2009
longitudinal
4
4
0
Design
Management
2,100
1,840
1,840
1,859
5
11
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.817
0.183
0.87
0.463
0.92
0.167
0.5
true
false
0.6
increase_investment
wait_and_test
true
Before declines, Management leans toward withdrawal while Design leans toward conflict. Before recoveries, the pattern shifts to Management leaning withdrawal and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:32:15.222190
4
[439, 271, 452, 678]
study1_longitudinal
Management_vs_Design
ES2010
longitudinal
4
4
0
Design
Management
1,511
1,389
1,389
1,408
6
23
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.995
0.005
0.81
0.405
0.88
0.091
0.111
false
false
0.55
reengage
reciprocate
true
Before declines, Management leans toward withdrawal while Design leans toward withdrawal. Before recoveries, the pattern shifts to Management leaning humor and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:32:26.606039
4
[132, 386, 468, 403]
study1_longitudinal
Management_vs_Design
ES2011
longitudinal
4
4
0
Design
Management
1,776
1,540
1,540
1,563
4
13
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.928
0.072
0.872
0.469
0.92
0.143
0.667
true
true
0.7
increase_investment
wait_and_test
true
Before declines, Management leans toward closing while Design leans toward vulnerability. Before recoveries, the pattern shifts to Management leaning power dynamics and Design leaning information exchange. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:32:39.041273
4
[234, 351, 435, 520]
study1_longitudinal
Management_vs_Design
ES2012
longitudinal
4
4
0
Design
Management
1,433
1,284
1,284
1,301
6
34
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.832
0.168
0.808
0.42
0.88
0.2
0
false
true
0.5
change_tone
continue_withdrawal
false
null
2026-07-23T00:32:50.574182
4
[142, 330, 521, 291]
study1_longitudinal
Management_vs_Design
ES2013
longitudinal
4
4
0
Design
Management
1,343
1,245
1,245
1,263
6
10
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.919
0.081
0.837
0.431
0.9
0.083
0.125
false
false
0.6
reengage
reciprocate
false
null
2026-07-23T00:33:00.558258
4
[121, 369, 376, 379]
study1_longitudinal
Management_vs_Design
ES2014
longitudinal
4
4
0
Design
Management
2,105
1,895
1,895
1,922
12
15
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.737
0.263
0.671
0.29
0.8
0.133
0.2
false
true
0.75
change_tone
continue_withdrawal
true
Before declines, Management leans toward deal progression while Design leans toward withdrawal. Before recoveries, the pattern shifts to Management leaning withdrawal and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:33:12.153182
4
[161, 479, 532, 723]
study1_longitudinal
Management_vs_Design
ES2015
longitudinal
4
4
0
Design
Management
2,541
2,285
2,285
2,308
4
44
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.817
0.183
0.872
0.471
0.92
0.091
0
false
false
0.55
reengage
reciprocate
true
Before declines, Management leans toward withdrawal while Design leans toward withdrawal. Before recoveries, the pattern shifts to Management leaning affection and Design leaning confidence signalling. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:33:25.501948
4
[210, 602, 628, 845]
study1_longitudinal
Management_vs_Design
ES2016
longitudinal
4
4
0
Design
Management
1,350
1,134
1,134
1,146
12
80
declining
low
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.777
0.223
0.697
0.325
0.82
0.214
0
false
true
0.7
change_tone
continue_withdrawal
true
Before declines, Management leans toward withdrawal while Design leans toward bonding. Before recoveries, the pattern shifts to Management leaning withdrawal and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:33:35.903216
4
[228, 274, 225, 407]
study1_longitudinal
Management_vs_Design
IS1000
longitudinal
4
4
0
Design
Management
2,532
2,371
2,371
2,393
2
2
gliding
low
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.926
0.525
0.953
0.138
0.148
false
false
0.55
reengage
reciprocate
false
null
2026-07-23T00:33:52.490411
4
[361, 557, 476, 977]
study1_longitudinal
Management_vs_Design
IS1001
longitudinal
4
4
0
Design
Management
1,482
1,314
1,314
1,327
4
36
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.998
0.002
0.872
0.488
0.92
0.1
0.1
false
false
0.5
reengage
reciprocate
false
No persistent cross-actor mismatch detected — Management and Design's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:34:02.834407
4
[185, 544, 324, 261]
study1_longitudinal
Management_vs_Design
IS1002
longitudinal
4
4
0
Design
Management
1,378
1,268
1,268
1,288
10
10
normalizing
high
quantum_live
volatile
medium
structural friction
r_structural
0
1
0
0.735
0.325
0.44
0.153
0.168
false
false
0.592
reengage
reciprocate
false
null
2026-07-23T00:34:13.385953
3
[453, 490, 325]
study1_longitudinal
Management_vs_Design
IS1003
longitudinal
4
4
0
Design
Management
2,319
2,083
2,083
2,107
4
18
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.815
0.185
0.874
0.466
0.92
0
0.091
false
false
0.45
reengage
reciprocate
false
null
2026-07-23T00:34:26.206701
4
[248, 381, 569, 885]
study1_longitudinal
Management_vs_Design
IS1004
longitudinal
4
4
0
Design
Management
1,822
1,645
1,645
1,672
3
38
gliding
medium
phase_plus_quantum
stable near peak
medium
structural friction
r_structural
0
0.7
0.3
0.904
0.479
0.94
0.125
0.083
false
false
0.4
reengage
reciprocate
false
No persistent cross-actor mismatch detected — Management and Design's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:34:39.877497
4
[77, 371, 553, 644]
study1_longitudinal
Management_vs_Design
IS1005
longitudinal
4
4
0
Design
Management
1,053
915
915
928
3
21
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.999
0.001
0.906
0.499
0.94
0
0
false
false
0.65
reengage
reciprocate
true
Before declines, Management leans toward withdrawal while Design leans toward withdrawal. Before recoveries, the pattern shifts to Management leaning withdrawal and Design leaning commitment. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:34:49.851011
3
[160, 397, 358]
study1_longitudinal
Management_vs_Design
IS1006
longitudinal
4
4
0
Design
Management
2,547
2,306
2,306
2,321
7
11
declining
medium
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.947
0.053
0.776
0.396
0.86
0.364
0.111
false
true
0.55
change_tone
continue_withdrawal
false
null
2026-07-23T00:35:05.020968
4
[322, 520, 571, 893]
study1_longitudinal
Management_vs_Design
IS1007
longitudinal
4
4
0
Design
Management
1,770
1,558
1,558
1,571
5
9
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.787
0.213
0.841
0.424
0.9
0
0.091
false
false
0.45
reengage
reciprocate
true
Before declines, Management leans toward bonding while Design leans toward bonding. Before recoveries, the pattern shifts to Management leaning corporate/admin and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:35:14.804890
4
[230, 328, 467, 533]
study1_longitudinal
Management_vs_Design
IS1008
longitudinal
4
4
0
Design
Management
1,136
979
979
997
2
38
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.916
0.084
0.936
0.487
0.96
0.167
0.125
false
false
0.6
reengage
reciprocate
true
Before declines, Management leans toward withdrawal while Design leans toward vulnerability. Before recoveries, the pattern shifts to Management leaning investor pitch and Design leaning vulnerability. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:35:23.098800
4
[145, 221, 266, 347]
study1_longitudinal
Management_vs_Design
IS1009
longitudinal
4
4
0
Design
Management
1,588
1,480
1,480
1,501
11
11
gliding
high
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.713
0.321
0.829
0.205
0.129
false
false
0.576
reengage
reciprocate
false
null
2026-07-23T00:35:34.301469
4
[217, 453, 290, 520]
study1_longitudinal
Management_vs_Design
TS3003
longitudinal
4
4
0
Design
Management
1,732
1,545
1,545
1,562
9
46
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.643
0.357
0.743
0.367
0.84
0.167
0.125
false
false
0.6
reengage
reciprocate
true
Before declines, Management leans toward bonding while Design leans toward withdrawal. Before recoveries, the pattern shifts to Management leaning information exchange and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:35:45.648007
4
[227, 336, 339, 643]
study1_longitudinal
Management_vs_Design
TS3004
longitudinal
4
4
0
Design
Management
2,876
2,655
2,655
2,684
6
16
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.886
0.114
0.81
0.42
0.88
0.5
0.214
true
true
0.3
increase_investment
continue_withdrawal
false
No persistent cross-actor mismatch detected — Management and Design's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:36:01.060946
4
[375, 625, 785, 870]
study1_longitudinal
Management_vs_Design
TS3005
longitudinal
4
4
0
Design
Management
3,644
3,321
3,321
3,353
5
33
declining
medium
phase_plus_quantum
declining
medium
structural friction
r_structural
0
0.892
0.108
0.967
0.608
1
0
0.143
true
false
0.3
increase_investment
reciprocate
true
Before declines, Management leans toward bonding while Design leans toward withdrawal. Before recoveries, the pattern shifts to Management leaning withdrawal and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:36:19.898315
4
[324, 787, 718, 1492]
study1_longitudinal
Management_vs_Design
TS3006
longitudinal
4
4
0
Design
Management
3,936
3,629
3,629
3,672
3
3
gliding
low
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.896
0.505
0.933
0.144
0.156
false
false
0.592
reengage
reciprocate
false
null
2026-07-23T00:36:38.692573
4
[468, 907, 971, 1283]
study1_longitudinal
Management_vs_Design
TS3007
longitudinal
4
4
0
Design
Management
2,990
2,609
2,609
2,650
10
48
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.898
0.102
0.736
0.344
0.84
0.444
0.182
false
true
0.45
change_tone
continue_withdrawal
false
null
2026-07-23T00:36:54.870823
4
[368, 489, 663, 1089]
study1_longitudinal
Management_vs_Design
TS3008
longitudinal
4
4
0
Design
Management
2,838
2,535
2,535
2,569
4
4
gliding
medium
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.876
0.461
0.923
0.151
0.188
false
false
0.61
reengage
reciprocate
false
null
2026-07-23T00:37:11.443190
4
[309, 584, 698, 944]
study1_longitudinal
Management_vs_Design
TS3009
longitudinal
4
4
0
Design
Management
3,700
3,253
3,253
3,301
7
35
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.894
0.106
0.805
0.41
0.88
0.154
0.429
true
false
0.65
increase_investment
wait_and_test
true
Before declines, Management leans toward withdrawal while Design leans toward reengagement. Before recoveries, the pattern shifts to Management leaning investor pitch and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:37:29.491300
4
[516, 860, 958, 919]
study1_longitudinal
Management_vs_Design
TS3010
longitudinal
4
4
0
Design
Management
1,295
1,071
1,071
1,081
5
5
gliding
medium
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.854
0.46
0.907
0.191
0.133
false
false
0.577
reengage
reciprocate
false
null
2026-07-23T00:37:38.069085
4
[117, 212, 389, 353]
study1_longitudinal
Management_vs_Design
TS3011
longitudinal
4
4
0
Design
Management
2,207
1,999
1,999
2,027
3
28
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.929
0.071
0.903
0.466
0.94
0.182
0.111
false
false
0.55
reengage
reciprocate
true
Before declines, Management leans toward bonding while Design leans toward bonding. Before recoveries, the pattern shifts to Management leaning withdrawal and Design leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:37:51.670869
4
[276, 570, 571, 582]
study1_longitudinal
Management_vs_Design
TS3012
longitudinal
4
4
0
Design
Management
2,061
1,927
1,927
1,946
2
83
declining
low
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.879
0.121
0.938
0.537
0.96
0.125
0.167
false
false
0.4
reengage
reciprocate
true
Before declines, Management leans toward withdrawal while Design leans toward information exchange. Before recoveries, the pattern shifts to Management leaning withdrawal and Design leaning confidence signalling. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:38:04.769828
3
[169, 910, 848]
study2_longitudinal
Project_Manager_vs_Team
ES2002
longitudinal
4
4
0
Project Manager
Team
2,498
2,340
2,340
2,367
7
7
declining
medium
phase_plus_quantum
declining
low
structural friction
r_structural
0
1
0
0.806
0.413
0.88
0.143
0.077
false
false
0.35
reengage
reciprocate
false
null
2026-07-23T00:43:22.231804
4
null
study2_longitudinal
Marketing_vs_Team
ES2002
longitudinal
4
4
0
Team
Marketing
2,498
2,340
2,340
2,368
6
11
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.834
0.166
0.808
0.413
0.88
0.143
0.077
false
false
0.35
reengage
reciprocate
false
null
2026-07-23T00:43:37.912569
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2003
longitudinal
4
4
0
Project Manager
Team
1,281
1,204
1,204
1,223
5
8
gliding
medium
phase_plus_quantum
stable near peak
medium
structural friction
r_structural
0
0.826
0.174
0.844
0.44
0.9
0.333
0
false
true
0.45
pull_back
continue_withdrawal
false
null
2026-07-23T00:43:49.834008
4
null
study2_longitudinal
Marketing_vs_Team
ES2003
longitudinal
4
4
0
Team
Marketing
1,281
1,204
1,204
1,218
5
8
gliding
medium
phase_plus_quantum
stable near peak
medium
structural friction
r_structural
0
0.966
0.034
0.844
0.44
0.9
0
0.214
true
false
0.3
increase_investment
reciprocate
false
null
2026-07-23T00:43:59.536501
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2004
longitudinal
4
4
0
Project Manager
Team
2,208
2,069
2,069
2,091
5
39
declining
medium
phase_plus_quantum
declining
medium
structural friction
r_structural
0
0.797
0.203
0.842
0.449
0.9
0.167
0.143
true
false
0.3
increase_investment
reciprocate
false
No persistent cross-actor mismatch detected — Project Manager and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:44:14.850284
4
null
study2_longitudinal
Marketing_vs_Team
ES2004
longitudinal
4
4
0
Team
Marketing
2,208
2,069
2,069
2,088
5
36
declining
medium
phase_plus_quantum
declining
medium
structural friction
r_structural
0
0.916
0.084
0.842
0.449
0.9
0
0.2
true
false
0.25
increase_investment
reciprocate
true
Before declines, Marketing leans toward bonding while Team leans toward withdrawal. Before recoveries, the pattern shifts to Marketing leaning withdrawal and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:44:30.400879
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2005
longitudinal
4
4
0
Project Manager
Team
2,160
1,901
1,901
1,928
2
2
gliding
low
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.938
0.524
0.96
0.179
0.163
true
false
0.341
increase_investment
reciprocate
false
null
2026-07-23T00:44:43.432478
4
null
study2_longitudinal
Marketing_vs_Team
ES2005
longitudinal
4
4
0
Team
Marketing
2,160
1,901
1,901
1,922
2
2
gliding
low
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.938
0.524
0.96
0.18
0.161
true
false
0.168
increase_investment
reciprocate
false
null
2026-07-23T00:44:56.140307
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2006
longitudinal
4
4
0
Project Manager
Team
2,432
2,182
2,182
2,207
8
36
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.879
0.121
0.8
0.394
0.88
0.667
0.357
true
true
0.3
increase_investment
continue_withdrawal
true
Before declines, Project Manager leans toward deal progression while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning conflict and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:45:10.313338
4
null
study2_longitudinal
Marketing_vs_Team
ES2006
longitudinal
4
4
0
Team
Marketing
2,432
2,182
2,182
2,202
8
36
stalled
medium
phase_plus_quantum
flat
medium
structural friction
r_structural
0
0.672
0.328
0.8
0.394
0.88
0.143
0.615
true
false
0.35
increase_investment
wait_and_test
true
Before declines, Marketing leans toward intimacy while Team leans toward withdrawal. Before recoveries, the pattern shifts to Marketing leaning withdrawal and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:45:22.651106
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2007
longitudinal
4
4
0
Project Manager
Team
1,708
1,501
1,501
1,527
7
27
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.782
0.218
0.803
0.393
0.88
0.143
0.308
true
false
0.35
increase_investment
reciprocate
false
No persistent cross-actor mismatch detected — Project Manager and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:45:34.330239
4
null
study2_longitudinal
Marketing_vs_Team
ES2007
longitudinal
4
4
0
Team
Marketing
1,708
1,501
1,501
1,520
7
7
declining
medium
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.844
0.156
0.803
0.393
0.88
0.286
0.231
false
false
0.35
reengage
reciprocate
false
null
2026-07-23T00:45:45.873876
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2008
longitudinal
4
4
0
Project Manager
Team
2,027
1,832
1,832
1,859
8
7
gliding
medium
phase_plus_quantum
stable near peak
medium
structural friction
r_structural
0
0.993
0.007
0.776
0.383
0.86
0.2
0.333
true
false
0.25
increase_investment
reciprocate
false
null
2026-07-23T00:46:01.900424
4
null
study2_longitudinal
Marketing_vs_Team
ES2008
longitudinal
4
4
0
Team
Marketing
2,027
1,832
1,832
1,851
8
7
stalled
medium
phase_plus_quantum
flat
low
structural friction
r_structural
0
0.742
0.258
0.773
0.383
0.86
0.375
0.25
false
false
0.4
reengage
reciprocate
false
null
2026-07-23T00:46:17.030779
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2009
longitudinal
4
4
0
Project Manager
Team
2,100
1,840
1,840
1,869
6
17
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.807
0.193
0.834
0.431
0.9
0.286
0.385
false
false
0.35
reengage
reciprocate
false
No persistent cross-actor mismatch detected — Project Manager and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:46:34.645815
4
null
study2_longitudinal
Marketing_vs_Team
ES2009
longitudinal
4
4
0
Team
Marketing
2,100
1,840
1,840
1,862
5
11
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.749
0.251
0.87
0.463
0.92
0
0.375
true
false
0.2
increase_investment
reciprocate
false
No persistent cross-actor mismatch detected — Marketing and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:46:49.650697
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2010
longitudinal
4
4
0
Project Manager
Team
1,511
1,389
1,389
1,411
6
22
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.791
0.209
0.81
0.405
0.88
0.167
0.071
true
false
0.3
increase_investment
reciprocate
true
Before declines, Project Manager leans toward rapport building while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning insecurity signals and Team leaning humor. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:47:02.398859
4
null
study2_longitudinal
Marketing_vs_Team
ES2010
longitudinal
4
4
0
Team
Marketing
1,511
1,389
1,389
1,409
8
23
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.807
0.193
0.773
0.374
0.86
0.167
0.143
true
false
0.3
increase_investment
reciprocate
true
Before declines, Marketing leans toward withdrawal while Team leans toward withdrawal. Before recoveries, the pattern shifts to Marketing leaning humor and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:47:14.168787
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2011
longitudinal
4
4
0
Project Manager
Team
1,776
1,540
1,540
1,559
4
9
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.884
0.116
0.872
0.469
0.92
0.286
0.308
false
false
0.35
reengage
reciprocate
false
null
2026-07-23T00:47:26.569435
4
null
study2_longitudinal
Marketing_vs_Team
ES2011
longitudinal
4
4
0
Team
Marketing
1,776
1,540
1,540
1,557
5
13
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.999
0.001
0.84
0.43
0.9
0
0.429
true
false
0.3
increase_investment
wait_and_test
true
Before declines, Marketing leans toward closing while Team leans toward withdrawal. Before recoveries, the pattern shifts to Marketing leaning power dynamics and Team leaning information exchange. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:47:39.644877
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2012
longitudinal
4
4
0
Project Manager
Team
1,433
1,284
1,284
1,305
6
34
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.843
0.157
0.808
0.42
0.88
0.2
0
false
true
0.5
change_tone
continue_withdrawal
false
null
2026-07-23T00:47:52.373228
4
null
study2_longitudinal
Marketing_vs_Team
ES2012
longitudinal
4
4
0
Team
Marketing
1,433
1,284
1,284
1,294
5
34
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.874
0.126
0.842
0.445
0.9
0.5
0.05
true
true
0
increase_investment
continue_withdrawal
false
null
2026-07-23T00:48:04.280322
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2013
longitudinal
4
4
0
Project Manager
Team
1,343
1,245
1,245
1,265
6
10
stalled
medium
phase_plus_quantum
flat
low
structural friction
r_structural
0
0.938
0.062
0.837
0.431
0.9
0.2
0.067
true
false
0.25
increase_investment
reciprocate
true
Before declines, Project Manager leans toward withdrawal while Team leans toward support. Before recoveries, the pattern shifts to Project Manager leaning investor pitch and Team leaning investor pitch. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:48:17.533119
4
null
study2_longitudinal
Marketing_vs_Team
ES2013
longitudinal
4
4
0
Team
Marketing
1,343
1,245
1,245
1,261
6
10
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.831
0.169
0.837
0.431
0.9
0
0.154
true
false
0.35
increase_investment
reciprocate
false
null
2026-07-23T00:48:30.863482
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2014
longitudinal
4
4
0
Project Manager
Team
2,105
1,895
1,895
1,916
13
15
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.761
0.239
0.638
0.255
0.78
0.286
0.154
false
false
0.35
follow_up
reciprocate
true
Before declines, Project Manager leans toward deal progression while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning withdrawal and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:48:47.176951
4
null
study2_longitudinal
Marketing_vs_Team
ES2014
longitudinal
4
4
0
Team
Marketing
2,105
1,895
1,895
1,925
12
13
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.578
0.422
0.672
0.286
0.8
0
0.25
true
false
0.4
increase_investment
reciprocate
true
Before declines, Marketing leans toward withdrawal while Team leans toward deal progression. Before recoveries, the pattern shifts to Marketing leaning withdrawal and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:49:02.973648
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2015
longitudinal
4
4
0
Project Manager
Team
2,541
2,285
2,285
2,301
4
45
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.919
0.081
0.872
0.471
0.92
0.143
0
false
false
0.35
reengage
reciprocate
true
Before declines, Project Manager leans toward withdrawal while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning affection and Team leaning confidence signalling. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:49:20.398346
4
null
study2_longitudinal
Marketing_vs_Team
ES2015
longitudinal
4
4
0
Team
Marketing
2,541
2,285
2,285
2,301
6
45
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.781
0.219
0.839
0.446
0.9
0
0.125
true
false
0.2
increase_investment
reciprocate
false
null
2026-07-23T00:49:39.274975
4
null
study2_longitudinal
Project_Manager_vs_Team
ES2016
longitudinal
4
4
0
Project Manager
Team
1,350
1,134
1,134
1,146
10
80
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.677
0.323
0.735
0.35
0.84
0.111
0.091
false
false
0.45
reengage
reciprocate
false
No persistent cross-actor mismatch detected — Project Manager and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:49:51.531233
4
null
study2_longitudinal
Marketing_vs_Team
ES2016
longitudinal
4
4
0
Team
Marketing
1,350
1,134
1,134
1,143
10
80
gliding
medium
phase_plus_quantum
stable near peak
medium
structural friction
r_structural
0
0.847
0.153
0.735
0.35
0.84
0.2
0.067
true
false
0.25
increase_investment
reciprocate
false
No persistent cross-actor mismatch detected — Marketing and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:50:03.344081
4
null
study2_longitudinal
Project_Manager_vs_Team
IS1000
longitudinal
4
4
0
Project Manager
Team
2,532
2,371
2,371
2,396
2
2
gliding
low
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.926
0.525
0.953
0.169
0.135
true
false
0.28
increase_investment
reciprocate
false
null
2026-07-23T00:50:20.746981
4
null
study2_longitudinal
Marketing_vs_Team
IS1000
longitudinal
4
4
0
Team
Marketing
2,532
2,371
2,371
2,398
2
2
gliding
low
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.926
0.525
0.953
0.117
0.153
true
false
0.274
increase_investment
reciprocate
false
null
2026-07-23T00:50:36.532515
4
null
study2_longitudinal
Project_Manager_vs_Team
IS1001
longitudinal
4
4
0
Project Manager
Team
1,482
1,314
1,314
1,330
4
36
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.947
0.053
0.872
0.488
0.92
0.111
0.091
false
false
0.45
reengage
reciprocate
false
No persistent cross-actor mismatch detected — Project Manager and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:50:48.932776
4
null
study2_longitudinal
Marketing_vs_Team
IS1001
longitudinal
4
4
0
Team
Marketing
1,482
1,314
1,314
1,330
4
36
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.821
0.179
0.872
0.488
0.92
0
0.105
true
false
0.05
increase_investment
reciprocate
true
Before declines, Marketing leans toward playful tension while Team leans toward withdrawal. Before recoveries, the pattern shifts to Marketing leaning information exchange and Team leaning affection. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:51:01.152341
4
null
study2_longitudinal
Project_Manager_vs_Team
IS1002
longitudinal
4
4
0
Project Manager
Team
1,378
1,268
1,268
1,282
10
10
normalizing
high
quantum_live
volatile
medium
structural friction
r_structural
0
1
0
0.735
0.325
0.44
0.145
0.161
true
false
0.344
reengage
reciprocate
false
null
2026-07-23T00:51:12.031057
3
null
study2_longitudinal
Marketing_vs_Team
IS1002
longitudinal
4
4
0
Team
Marketing
1,378
1,268
1,268
1,286
10
10
normalizing
high
quantum_live
volatile
medium
structural friction
r_structural
0
1
0
0.735
0.325
0.44
0.157
0.155
true
false
0.252
reengage
reciprocate
false
null
2026-07-23T00:51:25.976515
3
null
study2_longitudinal
Project_Manager_vs_Team
IS1003
longitudinal
4
4
0
Project Manager
Team
2,319
2,083
2,083
2,106
4
18
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.819
0.181
0.874
0.476
0.92
0
0.077
false
false
0.35
reengage
reciprocate
true
Before declines, Project Manager leans toward closing while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning confidence signalling and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:51:42.464330
4
null
study2_longitudinal
Marketing_vs_Team
IS1003
longitudinal
4
4
0
Team
Marketing
2,319
2,083
2,083
2,097
4
23
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.948
0.052
0.874
0.466
0.92
0
0.059
true
false
0.15
increase_investment
reciprocate
true
Before declines, Marketing leans toward withdrawal while Team leans toward closing. Before recoveries, the pattern shifts to Marketing leaning withdrawal and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:51:58.643054
4
null
study2_longitudinal
Project_Manager_vs_Team
IS1004
longitudinal
4
4
0
Project Manager
Team
1,822
1,645
1,645
1,669
3
49
normalizing
high
phase_plus_quantum
settling after peak
medium
structural friction
r_structural
0
1
0
0.904
0.479
0.94
0.167
0.071
true
false
0.3
increase_investment
reciprocate
true
Before declines, Project Manager leans toward withdrawal while Team leans toward bonding. Before recoveries, the pattern shifts to Project Manager leaning confidence signalling and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:52:13.980369
4
null
study2_longitudinal
Marketing_vs_Team
IS1004
longitudinal
4
4
0
Team
Marketing
1,822
1,645
1,645
1,676
3
49
normalizing
high
phase_plus_quantum
settling after peak
medium
structural friction
r_structural
0
0.797
0.203
0.904
0.479
0.94
0
0.111
true
false
0.1
increase_investment
reciprocate
false
null
2026-07-23T00:52:27.208968
4
null
study2_longitudinal
Project_Manager_vs_Team
IS1005
longitudinal
4
4
0
Project Manager
Team
1,053
915
915
931
4
21
normalizing
medium
phase_plus_quantum
settling after peak
medium
structural friction
r_structural
0
0.985
0.015
0.873
0.467
0.92
0.125
0
false
false
0.4
reengage
reciprocate
true
Before declines, Project Manager leans toward closing while Team leans toward withdrawal. Not enough recovery evidence yet to confirm what pairing precedes climbs instead.
2026-07-23T00:52:36.327725
3
null
study2_longitudinal
Marketing_vs_Team
IS1005
longitudinal
4
4
0
Team
Marketing
1,053
915
915
926
3
20
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.983
0.017
0.906
0.499
0.94
0
0
true
false
0.3
increase_investment
reciprocate
true
Before declines, Marketing leans toward withdrawal while Team leans toward withdrawal. Before recoveries, the pattern shifts to Marketing leaning withdrawal and Team leaning commitment. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:52:47.187922
3
null
study2_longitudinal
Project_Manager_vs_Team
IS1006
longitudinal
4
4
0
Project Manager
Team
2,547
2,306
2,306
2,327
7
11
declining
medium
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.727
0.273
0.776
0.396
0.86
0.286
0.231
false
false
0.35
reengage
reciprocate
false
null
2026-07-23T00:53:03.357592
4
null
study2_longitudinal
Marketing_vs_Team
IS1006
longitudinal
4
4
0
Team
Marketing
2,547
2,306
2,306
2,329
7
11
declining
medium
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.767
0.233
0.776
0.396
0.86
0.5
0.188
true
true
0.2
increase_investment
continue_withdrawal
false
null
2026-07-23T00:53:18.007638
4
null
study2_longitudinal
Project_Manager_vs_Team
IS1007
longitudinal
4
4
0
Project Manager
Team
1,770
1,558
1,558
1,570
5
12
declining
low
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.882
0.118
0.841
0.424
0.9
0
0.059
true
false
0.15
increase_investment
reciprocate
true
Before declines, Project Manager leans toward bonding while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning corporate/admin and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:53:31.840743
4
null
study2_longitudinal
Marketing_vs_Team
IS1007
longitudinal
4
4
0
Team
Marketing
1,770
1,558
1,558
1,576
5
9
declining
low
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.69
0.31
0.841
0.424
0.9
0
0.071
true
false
0.3
increase_investment
reciprocate
true
Before declines, Marketing leans toward bonding while Team leans toward bonding. Before recoveries, the pattern shifts to Marketing leaning confidence signalling and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:53:44.539851
4
null
study2_longitudinal
Project_Manager_vs_Team
IS1008
longitudinal
4
4
0
Project Manager
Team
1,136
979
979
994
2
32
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.716
0.284
0.936
0.487
0.96
0
0.2
true
false
0.25
increase_investment
reciprocate
true
Before declines, Project Manager leans toward withdrawal while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning support issue and Team leaning follow-up. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:53:56.013124
4
null
study2_longitudinal
Marketing_vs_Team
IS1008
longitudinal
4
4
0
Team
Marketing
1,136
979
979
1,000
2
38
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.913
0.087
0.936
0.491
0.96
0.286
0.077
false
true
0.35
pull_back
continue_withdrawal
true
Before declines, Marketing leans toward withdrawal while Team leans toward withdrawal. Before recoveries, the pattern shifts to Marketing leaning investor pitch and Team leaning confidence signalling. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:54:08.607698
4
null
study2_longitudinal
Project_Manager_vs_Team
IS1009
longitudinal
4
4
0
Project Manager
Team
1,588
1,480
1,480
1,499
11
11
gliding
high
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.713
0.321
0.829
0.218
0.153
true
false
0.3
increase_investment
reciprocate
false
null
2026-07-23T00:54:24.557201
4
null
study2_longitudinal
Marketing_vs_Team
IS1009
longitudinal
4
4
0
Team
Marketing
1,588
1,480
1,480
1,495
11
11
gliding
high
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.713
0.321
0.829
0.198
0.163
true
false
0.28
increase_investment
reciprocate
false
null
2026-07-23T00:54:38.303920
4
null
study2_longitudinal
Project_Manager_vs_Team
TS3003
longitudinal
4
4
0
Project Manager
Team
1,732
1,545
1,545
1,568
9
46
declining
low
phase_plus_quantum
declining
low
structural friction
r_structural
0
0.673
0.327
0.743
0.367
0.84
0.143
0.154
false
false
0.35
reengage
reciprocate
true
Before declines, Project Manager leans toward logistics while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning bonding and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:54:52.373555
4
null
study2_longitudinal
Marketing_vs_Team
TS3003
longitudinal
4
4
0
Team
Marketing
1,732
1,545
1,545
1,557
9
46
declining
medium
phase_plus_quantum
declining
low
structural friction
r_structural
0
1
0
0.743
0.367
0.84
0.2
0.133
true
false
0.25
increase_investment
reciprocate
true
Before declines, Marketing leans toward withdrawal while Team leans toward bonding. Before recoveries, the pattern shifts to Marketing leaning information exchange and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:55:04.931698
4
null
study2_longitudinal
Project_Manager_vs_Team
TS3004
longitudinal
4
4
0
Project Manager
Team
2,876
2,655
2,655
2,685
7
16
declining
low
phase_plus_quantum
declining
medium
structural friction
r_structural
0
0.992
0.008
0.806
0.42
0.88
0
0.316
true
false
0.05
increase_investment
reciprocate
false
No persistent cross-actor mismatch detected — Project Manager and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:55:26.026618
4
null
study2_longitudinal
Marketing_vs_Team
TS3004
longitudinal
4
4
0
Team
Marketing
2,876
2,655
2,655
2,675
6
17
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.808
0.192
0.81
0.42
0.88
0.6
0.2
true
true
0.25
increase_investment
continue_withdrawal
true
Before declines, Marketing leans toward withdrawal while Team leans toward withdrawal. Before recoveries, the pattern shifts to Marketing leaning resolution and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:55:46.503668
4
null
study2_longitudinal
Project_Manager_vs_Team
TS3005
longitudinal
4
4
0
Project Manager
Team
3,644
3,321
3,321
3,354
5
33
progressing
medium
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.847
0.153
0.967
0.608
1
0.5
0.1
true
true
0
increase_investment
continue_withdrawal
true
Before declines, Project Manager leans toward bonding while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning confidence signalling and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:56:17.683002
4
null
study2_longitudinal
Marketing_vs_Team
TS3005
longitudinal
4
4
0
Team
Marketing
3,644
3,321
3,321
3,358
5
33
stalled
medium
phase_plus_quantum
flat
medium
structural friction
r_structural
0
0.98
0.02
0.967
0.608
1
0
0.143
true
false
0.3
increase_investment
reciprocate
true
Before declines, Marketing leans toward withdrawal while Team leans toward withdrawal. Before recoveries, the pattern shifts to Marketing leaning value pitch and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:56:36.100187
4
null
study2_longitudinal
Project_Manager_vs_Team
TS3006
longitudinal
4
4
0
Project Manager
Team
3,936
3,629
3,629
3,674
3
3
gliding
low
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.896
0.505
0.933
0.13
0.158
true
false
0.287
increase_investment
reciprocate
false
null
2026-07-23T00:57:00.849449
4
null
study2_longitudinal
Marketing_vs_Team
TS3006
longitudinal
4
4
0
Team
Marketing
3,936
3,629
3,629
3,672
3
3
gliding
low
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.896
0.505
0.933
0.168
0.142
true
false
0.309
increase_investment
reciprocate
false
null
2026-07-23T00:57:23.684370
4
null
study2_longitudinal
Project_Manager_vs_Team
TS3007
longitudinal
4
4
0
Project Manager
Team
2,990
2,609
2,609
2,643
10
48
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.91
0.09
0.736
0.344
0.84
0.5
0.25
true
true
0.2
increase_investment
continue_withdrawal
false
No persistent cross-actor mismatch detected — Project Manager and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:57:42.940706
4
null
study2_longitudinal
Marketing_vs_Team
TS3007
longitudinal
4
4
0
Team
Marketing
2,990
2,609
2,609
2,645
10
26
critical_decline
low
phase_plus_quantum
critical decline
low
structural friction
r_structural
0
0.817
0.183
0.736
0.344
0.84
0.4
0.267
true
false
0.25
increase_investment
reciprocate
false
null
2026-07-23T00:58:01.804117
4
null
study2_longitudinal
Project_Manager_vs_Team
TS3008
longitudinal
4
4
0
Project Manager
Team
2,838
2,535
2,535
2,588
5
5
gliding
medium
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.865
0.45
0.917
0.178
0.164
true
false
0.343
increase_investment
reciprocate
false
null
2026-07-23T00:58:23.001977
4
null
study2_longitudinal
Marketing_vs_Team
TS3008
longitudinal
4
4
0
Team
Marketing
2,838
2,535
2,535
2,574
4
4
gliding
medium
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.873
0.452
0.922
0.13
0.184
true
false
0.272
increase_investment
reciprocate
false
null
2026-07-23T00:58:40.897236
4
null
study2_longitudinal
Project_Manager_vs_Team
TS3009
longitudinal
4
4
0
Project Manager
Team
3,700
3,253
3,253
3,293
8
28
progressing
high
phase_plus_quantum
ascending
medium
structural friction
r_structural
0
0.913
0.087
0.77
0.385
0.86
0.286
0.308
false
false
0.35
reengage
reciprocate
true
Before declines, Project Manager leans toward withdrawal while Team leans toward withdrawal. Before recoveries, the pattern shifts to Project Manager leaning commitment and Team leaning withdrawal. This system's own history shows decline tracks this specific combination, not either category alone.
2026-07-23T00:59:01.876849
4
null
study2_longitudinal
Marketing_vs_Team
TS3009
longitudinal
4
4
0
Team
Marketing
3,700
3,253
3,253
3,291
7
35
declining
medium
phase_plus_quantum
declining
medium
structural friction
r_structural
0
0.801
0.199
0.805
0.41
0.88
0.167
0.286
true
false
0.3
increase_investment
reciprocate
false
No persistent cross-actor mismatch detected — Marketing and Team's dominant emissions before declines and recoveries don't diverge meaningfully.
2026-07-23T00:59:22.182448
4
null
study2_longitudinal
Project_Manager_vs_Team
TS3010
longitudinal
4
4
0
Project Manager
Team
1,295
1,071
1,071
1,083
5
5
gliding
medium
quantum_live
stable
medium
structural friction
r_structural
0
1
0
0.854
0.46
0.907
0.217
0.139
false
false
0.387
reengage
reciprocate
false
null
2026-07-23T00:59:32.306146
4
null
End of preview.

Hermion QMSum Interaction Dynamics

385 intelligence runs · 35 corporate product teams · 3 studies · 0 messages read

Dataset Description

This dataset contains structured interaction dynamics extracted from 35 real corporate product teams in the QMSum meeting corpus, processed through Hermion's physics-based interaction intelligence engine.

What makes this dataset novel: Every result was produced without reading message content. Raw transcripts were converted to encoded mathematical signals on the researcher's machine using the Hermion open source classifier. Only signals reached the intelligence engine. No message text was transmitted or stored.

This is the first structured dataset of interaction dynamics extracted from real corporate meeting data without semantic content access. LLM-based approaches on the same QMSum text produce unstructured, non-deterministic, non-comparable outputs. Every row in this dataset is the result of a live, successful API call — deterministic, structured, and reproducible.

Dataset Structure

Three CSV files, one per study:

summary_study1_longitudinal.csv — 35 rows

Management (Project Manager + Marketing) vs Design (Industrial Designer + User Interface) across the full longitudinal arc of each team.

summary_study2_longitudinal.csv — 70 rows

Project Manager solo vs Team (35 runs) and Marketing solo vs Team (35 runs). Measures structural influence of each executive role against the collective team.

summary_study3_longitudinal.csv — 280 rows

Eight executive-anchored pairings per team. All pairings where PM or Marketing is one side. Covers every combination from the executive dyad to individual role pairs to executive vs full team.

Fields

Field Type Description
meeting_id string Project team identifier (e.g. ES2004, TS3012)
run_label string Pairing type (e.g. PM_vs_UI, Management_vs_Design)
phase string longitudinal (all studies use full project arc)
session_count int Number of sessions concatenated (3 or 4)
progression_pct float Current interaction progression percentage
peak_pct float Peak progression reached across project arc
momentum string progressing / declining / critical_decline / gliding / stalled / normalizing
interest_level string high / medium / low
topology_mode string phase_plus_quantum / quantum_live
topology_label string ascending / declining / stable / critical decline / volatile / stable near peak
primary_resistance string structural friction (universal across all 35 teams)
dominant_component string r_structural / r_temporal / r_asymmetric
structural_share float Share of resistance that is structural (0–1)
asymmetric_share float Share of resistance that is asymmetric (0–1)
ghosting_risk float Risk score 0–1
rejection_risk float Risk score 0–1
stall_risk float Risk score 0–1
self_positive_ratio float Positive signal ratio for self actor
other_positive_ratio float Positive signal ratio for other actor
self_withdrawing bool Whether self actor shows withdrawal pattern
other_withdrawing bool Whether other actor shows withdrawal pattern
self_signal_share float Share of total signals from self actor
next_move string Recommended action for self actor
other_next_move string Predicted next action for other actor
mismatch_detected bool Whether cross-actor divergence was detected
mismatch_desc string Description of divergence pattern (if detected)
recovery_capacity string high / medium / low
group_a string Group A label
group_b string Group B label
signal_count int Number of encoded signals processed
prepared_messages int Messages after window algorithm filtering

Key Findings

  1. PM structural amplification: Project Manager produces +10.5pp higher peak dynamics than equivalent Marketing pairings on average. Gap is largest in PM/UI dyad (+13.1pp).

  2. PM/UI dyad: The direct relationship between Project Manager and User Interface designer produces the highest average peak of any pairing (25.5%) — above the executive pair, above PM vs full team.

  3. Peak lifecycle: Teams average 4.2x between peak and current progression at project conclusion. Peak clustering confirmed in development phase (5.6% avg vs 4.9% kick-off and 5.1% conclusion).

  4. Universal structural friction: 100% of 35 teams show structural friction as primary resistance type. Structural share averages 87.6%.

  5. Pre-decline signal: 48% of teams show detectable cross-actor mismatch before decline events. Pattern is consistent — Management leans toward closing/deal-progression while Design leans toward withdrawal.

  6. Non-linear coupling: Team IS1007 — PM vs combined Design peaked at 85% while PM vs ID alone peaked at 33% and PM vs UI alone at 45%. The combination unlocked a peak neither individual produced.

Methodology

Dataset source: QMSum — Product domain, 35 teams with consistent 4-role labelling (Project Manager, Marketing, Industrial Designer, User Interface).

Longitudinal design: All sessions per team concatenated chronologically (kick-off → development → conclusion). One continuous interaction arc per team. Session counts range from 3 to 4. Total turns per team range from 915 to 3,936. No arc sampling required.

Window algorithm: When actors are isolated in pairings, only messages within ±N positions of the other group in the original stream are retained. N = min(max(groupSpacingA, groupSpacingB), 20). This isolates genuine exchanges rather than treating all messages as relevant to the pairing.

Privacy: Raw QMSum transcripts were processed locally. No message text was transmitted to the Hermion engine. The classifier encodes messages into mathematical signals on the researcher's machine.

Pipeline: Fully open source at github.com/hermionai/hermion-research. Reproducible with a Hermion API key.

Reproduce These Results 

git clone https://github.com/hermionai/hermion-research
cd hermion-research
pip install -r requirements.txt
python data/download.py
export HERMION_API_KEY=herm_sk_live_...
python studies/study1_longitudinal.py
python studies/study2_longitudinal.py
python studies/study3_longitudinal.py

Get an API key at hermionai.xyz/get-started.

Citation

If you use this dataset, please cite:

@misc{hermion_qmsum_2026,
  title  = {Hermion QMSum Interaction Dynamics: Physics-Based Interaction
             Intelligence Across 35 Corporate Product Teams},
  author = {Egie, Mario},
  year   = {2026},
  url    = {https://huggingface.co/datasets/hermionai/hermion-qmsum-interaction-dynamics},
  note   = {Pipeline: github.com/hermionai/hermion-research}
}

Also cite the QMSum dataset:

@inproceedings{zhong-etal-2021-qmsum,
  title     = {QMSum: A New Benchmark for Query-based Multi-domain Meeting Summarization},
  author    = {Zhong, Ming and others},
  booktitle = {Proceedings of NAACL 2021},
  year      = {2021},
  url       = {https://github.com/Yale-LILY/QMSum}
}

Related Resources

License

CC BY 4.0 — use freely with attribution.

The underlying QMSum dataset is separately licensed. See Yale-LILY/QMSum for terms.

Hermion autonomously models the dynamics of relationships — making them descriptive, diagnosable, and predictive — without reading raw messages. Based on a proprietary physics model.

You cannot understand a relationship by violating it.
Downloads last month
72
Total file size:
457 kB