SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.19392 · 2024

ReXTime: A Benchmark Suite for Reasoning-Across-Time in Videos

arXiv · PDF · Open in the Atlas

Code that ran

5 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2406.19392.json · badge: /api/ran/2406.19392.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
rextime/rextime canonical (classifier label)
classifier basis: census
5 ran · 1 no recorded run
FunctionStatusWhere it lives
compute_gqa_accuracy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
rextime/rextime/evaluation/utils.py
code served (permissive licence) · get_code("ffb60af327ed779e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create sample data for the function arguments
pred_ans = ["cat", "dog", "bird", "cat", "dog"]
gt_ans = ["cat", "dog", "bird", "dog", "dog"]
pred_gt_iou = [0.9, 0.6, 0.4, 0.2, 0.8]
thres = [0.3, 0.5, 0.7]

ARGS = {
    "pred_ans": pred_ans,
    "gt_ans": gt_ans,
    "pred_gt_iou": pred_gt_iou,
    "thres": thres
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
compute_temporal_iou_batch_paired Ran
ran, honoured the contract we drafted
rextime/rextime/evaluation/utils.py
code served (permissive licence) · get_code("7069f9c03e6dffb9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create small sample data for temporal windows
# Each window is represented as [start_time, end_time]
# Create N=2 paired windows for testing

# Predicted windows
pred_windows = np.array([
    [1.0, 5.0],
    [2.0, 7.0]
], dtype=np.float32)

# Ground truth windows
gt_windows = np.array([
    [2.0, 6.0],
    [1.5, 8.0]
], dtype=np.float32)

ARGS = {
    "pred_windows": pred_windows,
    "gt_windows": gt_windows
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
decode_video Ran
ran; our contract draft was wrong, not the code
rextime/rextime/demo/request.py
code served (permissive licence) · get_code("8cd2e117730e89b6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import cv2
import tempfile
import os

# Create a small test video file
def create_test_video(filepath, num_frames=10, fps=30, width=320, height=240):
    """Create a simple test video file"""
    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
    out = cv2.VideoWriter(filepath, fourcc, fps, (width, height))
    
    for i in range(num_frames):
        # Create a simple frame with changing color
        frame = np.zeros((height, width, 3), dtype=np.uint8)
        frame[:, :] = (i * 25 % 256, i * 50 % 256, i * 100 % 256)
        out.write(frame)
    
    out.release()

# Create temporary video file
temp_dir = tempfile.gettempdir()
video_path = os.path.join(temp_dir, "test_video.mp4")
create_test_video(video_path, num_frames=100, fps=30)

ARGS = {
    "video_path": video_path,
    "total_frames": 50
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
extract_time Ran
ran, honoured the contract we drafted
rextime/rextime/demo/inference.py
code served (permissive licence) · get_code("fdb15050f1e957b1")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

from datetime import datetime

ARGS = {
    "response": "14:30:45 to 16:45:30"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_jsonl Ran
ran; our contract draft was wrong, not the code
rextime/rextime/evaluation/utils.py
code served (permissive licence) · get_code("f1d1cccccf038785")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import json
import tempfile
import os

# Create a temporary JSONL file with valid data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.jsonl")

# Write sample JSONL data
with open(temp_file, 'w') as f:
    f.write(json.dumps({"key": "value1", "number": 1}) + "\n")
    f.write(json.dumps({"key": "value2", "number": 2}) + "\n")
    f.write(json.dumps({"key": "value3", "number": 3}) + "\n")

ARGS = {
    "filename": temp_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_data_by_range Not yet run
unverified (harvested, no recorded run)
rextime/rextime/evaluation/rextime_eval.py
code served (permissive licence) · get_code("4f116c6e550636ff")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 1 the same month, 0 later) and 4 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
load_jsonl 8 other papers
7 dated earlier, 1 the same month
2001.09099 (2020-01)
TVR: A Large-Scale Dataset for Video-Subtitle Moment Retrieval
2107.06383 (2021-07)
How Much Can CLIP Benefit Vision-and-Language Tasks?
2022.findings-emnlp.24 (2022 (year only))
AssistSR: Task-oriented Video Segment Retrieval for Personal AI Assistant
and 5 more not named here
compute_temporal_iou_batch_paired 1 other paper
1 dated earlier
2403.02076 (2024-03)
VTG-GPT: Tuning-Free Zero-Shot Video Temporal Grounding with GPT

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We introduce ReXTime, a benchmark designed to rigorously test AI models' ability to perform temporal reasoning within video events. Specifically, ReXTime focuses on reasoning across time, i.e. human-like understanding when the question and its corresponding answer occur in different video segments. This form of reasoning, requiring advanced understanding of cause-and-effect relationships across video segments, poses significant challenges to even the frontier multimodal large language models. To facilitate this evaluation, we develop an automated pipeline for generating temporal reasoning question-answer pairs, significantly reducing the need for labor-intensive manual annotations. Our benchmark includes 921 carefully vetted validation samples and 2,143 test samples, each manually curated for accuracy and relevance. Evaluation results show that while frontier large language models outperform academic models, they still lag behind human performance by a significant 14.3% accuracy gap. Additionally, our pipeline creates a training dataset of 9,695 machine generated samples without manual effort, which empirical studies suggest can enhance the across-time reasoning via fine-tuning.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2406.19392")
get_code_for_paper("2406.19392")
have("2406.19392")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2406.19392.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2406.19392.svg)](https://syntology.ai/paper/2406.19392)

Connect an agent — have() is free.