SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.17868 · ICML · 2026

VidLaDA: Bidirectional Diffusion Large Language Models for Efficient Video Understanding

Zuxuan Wu, Weiyao Lin, Tieyuan Chen, Chaofan Gan, Zhihao He, Ziran Qin, Shijie Li, Kangyu Wang, Yang Shao

arXiv · PDF · Open in the Atlas

Code that ran

8 of 13 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 7 failures of our instrument; 5 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, measured 2026-11-22; 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/2601.17868.json · badge: /api/ran/2601.17868.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
ziHoHe/VidLaDA canonical (classifier label)
classifier basis: census
8 ran · 5 no recorded run
FunctionStatusWhere it lives
repeat_kv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ziHoHe/VidLaDA/train/llada_v_prepare/files/modeling_llada.py
code served (permissive licence) · get_code("30d7eec482ebf6b1")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 4, dtype family float, shape (2, 6, 5, 4) as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-04T23:09:20Z). 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 torch

torch.manual_seed(0)

# Create a tensor with shape (batch, num_key_value_heads, seqlen, head_dim)
# Using distinct sizes: batch=2, num_key_value_heads=3, seqlen=5, head_dim=4
hidden_states = torch.randn(2, 3, 5, 4)

# n_rep is the repetition factor
n_rep = 2

ARGS = {
    "hidden_states": hidden_states,
    "n_rep": n_rep,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 6, 5, 4) — shape (2, 6, 5, 4) 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
apply_rotary_pos_emb Ran
ran; our contract draft was wrong, not the code
ziHoHe/VidLaDA/train/llada_v_prepare/files/modeling_llada.py
code served (permissive licence) · get_code("bac65c3dafaec040")
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) and returned tuple, ndim 5, 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 torch

torch.manual_seed(0)

# Create small tensors for query and key
# Shape: [batch_size, heads, seq_len, head_dim]
batch_size = 2
heads = 3
seq_len = 4
head_dim = 8

q = torch.randn(batch_size, heads, seq_len, head_dim)
k = torch.randn(batch_size, heads, seq_len, head_dim)

# Create cos and sin tensors
# Shape: [batch_size, seq_len, head_dim]
cos = torch.randn(batch_size, seq_len, head_dim)
sin = torch.randn(batch_size, seq_len, head_dim)

# position_ids is optional and unused according to docstring
position_ids = None

# unsqueeze_dim defaults to 1
unsqueeze_dim = 1

ARGS = {
    "q": q,
    "k": k,
    "cos": cos,
    "sin": sin,
    "position_ids": position_ids,
    "unsqueeze_dim": unsqueeze_dim,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 5, dtype family float — a tuple 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
convert_to_scalar Ran
ran; our contract draft was wrong, not the code
ziHoHe/VidLaDA/train/trl/core.py
code served (permissive licence) · get_code("c32b636f13ed9ee5")
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)

ARGS = {
    "stats": {
        "loss": torch.tensor(0.5),
        "accuracy": torch.tensor([0.95]),
        "batch_size": 32,
        "learning_rate": 0.001,
        "numpy_metric": np.array(0.87),
        "numpy_array_metric": np.array([0.92]),
        "string_value": "test",
        "multi_dim_tensor": torch.randn(2, 3),
    }
}
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
extract_patches Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ziHoHe/VidLaDA/train/llava/mm_utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from HumanMLLM/LLaVA-Scissor, is pointer only (licence: NONE) · get_code("b0a851cae92754ff")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from HumanMLLM/LLaVA-Scissor, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

find_closest_simple_ratio Ran
ran, honoured the contract we drafted
ziHoHe/VidLaDA/train/create_mix_data.py
code served (permissive licence) · get_code("21c36f0bbf177093")
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, 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.

import torch
import numpy as np

# The function find_closest_simple_ratio takes:
# - target_ratio: a float representing the target ratio to match
# - max_val: an integer for the maximum value in the ratio (default 10)

# Looking at the function body:
# - target_ratio is compared with <= 0, divided, and used in arithmetic
# - max_val is used in range() calls
# Both parameters are simple numeric types.

# Set up test values
target_ratio = 1.618  # golden ratio as an example
max_val = 10

ARGS = {
    "target_ratio": target_ratio,
    "max_val": max_val,
}
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 tuple, ndim 1, dtype family int — a tuple 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
rename_keys_in_state_dict Ran
ran; our contract draft was wrong, not the code
ziHoHe/VidLaDA/train/llada_v_prepare/rename_checkpoint.py
code served (permissive licence) · get_code("d99baa5425e99407")
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 torch.nn as nn

torch.manual_seed(0)

# Create a minimal state_dict_shard with various key patterns
state_dict_shard = {
    # Keys that match the base_mapping_source_to_target_suffix pattern
    'model.transformer.blocks.0.q_proj.weight': torch.randn(4, 3),
    'model.transformer.blocks.0.k_proj.weight': torch.randn(4, 3),
    'model.transformer.blocks.0.v_proj.weight': torch.randn(4, 3),
    'model.transformer.blocks.0.attn_out.weight': torch.randn(4, 3),
    'model.transformer.blocks.0.ff_proj.weight': torch.randn(4, 3),
    'model.transformer.blocks.0.up_proj.weight': torch.randn(4, 3),
    'model.transformer.blocks.0.ff_out.weight': torch.randn(4, 3),
    'model.transformer.blocks.0.attn_norm.weight': torch.randn(4, 3),
    'model.transformer.blocks.0.post_attention_layernorm.weight': torch.randn(4, 3),
    'model.transformer.blocks.1.q_proj.weight': torch.randn(5, 2),
    'model.transformer.blocks.1.ff_norm.weight': torch.randn(5, 2),
    # Keys that match additional_direct_mapping
    'model.transformer.wte.weight': torch.randn(6, 7),
    'model.transformer.ln_f.weight': torch.randn(8, 9),
    'model.transformer.ff_out.weight': torch.randn(10, 11),
    # Key that doesn't match any mapping (should be discarded)
    'model.transformer.unknown_key.weight': torch.randn(3, 3),
}

ARGS = {
    'state_dict_shard': state_dict_shard,
    'save_bf16': False,
    'shard_name_for_debug': 'test_shard_0',
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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
rotate_half Ran
ran; our contract draft was wrong, not the code
ziHoHe/VidLaDA/train/llada_v_prepare/files/modeling_llada.py
code served (permissive licence) · get_code("b99eea6376d1e212")
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) and returned Tensor, ndim 2, dtype family float, shape (2, 4) 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 torch

torch.manual_seed(0)

# The function rotate_half takes a single parameter 'x' which is a tensor
# It accesses x.shape[-1] to get the last dimension
# It uses indexing with [..., :] slicing notation
# It calls torch.cat() to concatenate tensors

# Create a small tensor with shape (2, 4) - batch size 2, hidden dim 4
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
unpad_image Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ziHoHe/VidLaDA/train/llava/model/llava_arch.py
code served (permissive licence) · get_code("7606525af238fb64")
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) and returned Tensor, ndim 3, dtype family float, shape (3, 8, 6) 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 torch

torch.manual_seed(0)

# Create a padded image tensor in CxHxW format
# C=3 (RGB), H=8 (current height), W=6 (current width)
tensor = torch.randn(3, 8, 6)

# Original size as (width, height)
# Using different dimensions to be distinct: width=4, height=5
original_size = (4, 5)

ARGS = {
    "tensor": tensor,
    "original_size": original_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (3, 8, 6) — shape (3, 8, 6) 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
auto_pad_images Not yet run
unverified (harvested, no recorded run)
ziHoHe/VidLaDA/train/llava/mm_utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from HumanMLLM/LLaVA-Scissor, is pointer only (licence: NONE) · get_code("3c8823948724e269")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from HumanMLLM/LLaVA-Scissor, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

flatten_dict Not yet run
unverified (harvested, no recorded run)
ziHoHe/VidLaDA/train/trl/core.py
code served (permissive licence) · get_code("a51ae45a1f6c0b5b")
How it ran

no invocation recorded.

resize_and_center_crop Not yet run
unverified (harvested, no recorded run)
ziHoHe/VidLaDA/train/llava/mm_utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from HumanMLLM/LLaVA-Scissor, is pointer only (licence: NONE) · get_code("c80358dff96c25ce")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from HumanMLLM/LLaVA-Scissor, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

safe_load_tokenizer Not yet run
unverified (harvested, no recorded run)
ziHoHe/VidLaDA/train/llava/conversation.py
code served (permissive licence) · get_code("b20705b0dfd77ffc")
How it ran

no invocation recorded.

top_k_top_p_filtering Not yet run
unverified (harvested, no recorded run)
ziHoHe/VidLaDA/train/trl/core.py
code served (permissive licence) · get_code("2c3da386578af970")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 11 are also present in other papers whose code we hold (11 with at least one paper dated earlier, 10 the same month, 10 later) and 2 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
rotate_half 360 other papers
316 dated earlier, 7 the same month, 34 later, 3 undated · 30 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2203.08913 (2022-03)
Memorizing Transformers
2302.13971 (2023-02)
LLaMA: Open and Efficient Foundation Language Models
and 357 more not named here
repeat_kv 243 other papers
215 dated earlier, 6 the same month, 20 later, 2 undated · 22 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2203.08913 (2022-03)
Memorizing Transformers
2305.17888 (2023-05)
LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
and 240 more not named here
apply_rotary_pos_emb 94 other papers
78 dated earlier, 3 the same month, 10 later, 3 undated · 6 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2305.17888 (2023-05)
LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
2309.11674 (2023-09)
A Paradigm Shift in Machine Translation: Boosting Translation Performance of Large Language Models
and 91 more not named here
unpad_image 69 other papers
66 dated earlier, 3 later · 5 only as an assembly brick
2304.08485 (2023-04)
Visual Instruction Tuning
2309.09958 (2023-09)
An Empirical Study of Scaling Instruct-Tuned Large Multimodal Models
2310.03744 (2023-10)
Improved Baselines with Visual Instruction Tuning
and 66 more not named here
top_k_top_p_filtering 21 other papers
19 dated earlier, 1 the same month, 1 later
2305.12191 (2023-05)
Pointwise Mutual Information Based Metric and Decoding Strategy for Faithful Generation in Document Grounded Dialogs
2305.18290 (2023-05)
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
2311.17911 (2023-11)
OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocation
and 18 more not named here
extract_patches 20 other papers
18 dated earlier, 1 the same month, 1 later
2411.10332 (2024-11)
Number it: Temporal Grounding Videos like Flipping Manga
2411.11360 (2024-11)
CCExpert: Advancing MLLM Capability in Remote Sensing Change Captioning with Difference-Aware Integration and a Foundational Dataset
2411.14432 (2024-11)
Insight-V: Exploring Long-Chain Visual Reasoning with Multimodal Large Language Models
and 17 more not named here
auto_pad_images 19 other papers
17 dated earlier, 1 the same month, 1 later
2411.10332 (2024-11)
Number it: Temporal Grounding Videos like Flipping Manga
2411.11360 (2024-11)
CCExpert: Advancing MLLM Capability in Remote Sensing Change Captioning with Difference-Aware Integration and a Foundational Dataset
2411.14432 (2024-11)
Insight-V: Exploring Long-Chain Visual Reasoning with Multimodal Large Language Models
and 16 more not named here
convert_to_scalar 18 other papers
16 dated earlier, 1 the same month, 1 later
2305.18290 (2023-05)
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
2406.11280 (2024-06)
ISR-DPO: Aligning Large Multimodal Models for Videos by Iterative Self-Retrospective DPO
2406.19371 (2024-06)
Suri: Multi-constraint Instruction Following for Long-form Text Generation
and 15 more not named here
flatten_dict 18 other papers
16 dated earlier, 1 the same month, 1 later
2305.18290 (2023-05)
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
2406.11280 (2024-06)
ISR-DPO: Aligning Large Multimodal Models for Videos by Iterative Self-Retrospective DPO
2406.19371 (2024-06)
Suri: Multi-constraint Instruction Following for Long-form Text Generation
and 15 more not named here
resize_and_center_crop 18 other papers
16 dated earlier, 1 the same month, 1 later
2411.10332 (2024-11)
Number it: Temporal Grounding Videos like Flipping Manga
2411.11360 (2024-11)
CCExpert: Advancing MLLM Capability in Remote Sensing Change Captioning with Difference-Aware Integration and a Foundational Dataset
2411.14432 (2024-11)
Insight-V: Exploring Long-Chain Visual Reasoning with Multimodal Large Language Models
and 15 more not named here
safe_load_tokenizer 12 other papers
11 dated earlier, 1 the same month
2411.15024 (2024-11)
DyCoke: Dynamic Compression of Tokens for Fast Video Large Language Models
2412.03248 (2024-12)
AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning
2412.07689 (2024-12)
RoboTron-Drive: All-in-One Large Multimodal Model for Autonomous Driving
and 9 more not named here

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

Current Video Large Language Models (Video LLMs) typically encode frames via a vision encoder and employ an autoregressive (AR) LLM for understanding and generation. However, this AR paradigm inevitably faces a dual efficiency bottleneck: strictly unidirectional attention compromises understanding efficiency by hindering global spatiotemporal aggregation, while serial decoding restricts generation efficiency. To address this, we propose VidLaDA, a Video LLM based on Diffusion Language Models (DLMs) that leverages bidirectional attention to unlock comprehensive spatiotemporal modeling and decode tokens in parallel. To further mitigate the computational overhead of diffusion decoding, we introduce MARS-Cache, an acceleration strategy that prunes redundancy by combining asynchronous visual cache refreshing with frame-wise chunk attention. Experiments show VidLaDA rivals state-of-the-art AR baselines (e.g., Qwen2.5-VL and LLaVA-Video) and outperforms DLM baselines, with MARS-Cache delivering over 12× speedup without compromising accuracy. Code and checkpoints are open-sourced at https: //github.com/ziHoHe/VidLaDA.

For agents

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

get_harvested_code_for_paper("2601.17868")
get_code_for_paper("2601.17868")
have("2601.17868")

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

curl https://syntology.ai/api/ran/2601.17868.json

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

[![Syntology run record](https://syntology.ai/api/ran/2601.17868.svg)](https://syntology.ai/paper/2601.17868)

Connect an agent — have() is free.