SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.26110 · 2026

PRISM: A Plug-in Reproducible Infrastructure for Scalable Multimodal Continual Instruction Tuning

Da-Wei Zhou, Zhen-Hao Xie, Jun-Tao Tang, Yu-Cheng Shi

arXiv · PDF · Open in the Atlas

Code that ran

1 of 3 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; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 2 no recorded run), 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/2605.26110.json · badge: /api/ran/2605.26110.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
LAMDA-CL/Prism canonical (classifier label)
classifier basis: census
1 ran · 2 no recorded run
FunctionStatusWhere it lives
resolve_peft_target_spec_to_allowed_suffixes Ran
ran; our contract draft was wrong, not the code
LAMDA-CL/Prism/PEFT/utils/peft_target_modules.py
code served (permissive licence) · get_code("0f4f89772545d944")
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 set. 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 json
from typing import Any, List, Optional, Set

# First, define the PEFT_TARGET_MODULE_PRESETS that the function references
PEFT_TARGET_MODULE_PRESETS = {
    'attention': ['q_proj', 'k_proj', 'v_proj', 'out_proj'],
    'ffn': ['fc1', 'fc2', 'up_proj', 'down_proj'],
    'all': ['q_proj', 'k_proj', 'v_proj', 'out_proj', 'fc1', 'fc2', 'up_proj', 'down_proj'],
    'mlp': ['fc1', 'fc2'],
    'qkv': ['q_proj', 'k_proj', 'v_proj'],
}

# Test cases for different input types
test_cases = [
    None,  # None input - should return None
    ['q_proj', 'v_proj'],  # List of strings
    'attn',  # String preset
    'attn_qv',  # Special attention preset
    'attn_and_ffn',  # Combined preset
    'all',  # All modules
    '["q_proj", "v_proj"]',  # JSON string
    'q_proj, v_proj',  # Comma-separated string
    'q_proj  v_proj',  # Space-separated string (multiple spaces)
    'custom_module',  # Single custom module name
    ('q_proj', 'k_proj'),  # Tuple input
    {'q_proj', 'v_proj'},  # Set input
    frozenset(['q_proj', 'v_proj']),  # Frozenset input
]

# Create ARGS with one representative case
ARGS = {
    'raw': ['q_proj', 'v_proj', 'fc1']
}
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 set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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
load_tokenizer Not yet run
unverified (harvested, no recorded run)
LAMDA-CL/Prism/backbone/shared/model_loading.py
code served (permissive licence) · get_code("0f0205bc9e1110e2")
How it ran

no invocation recorded.

setup_quantization Not yet run
unverified (harvested, no recorded run)
LAMDA-CL/Prism/backbone/shared/model_loading.py
code served (permissive licence) · get_code("7fc1082f418aa422")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 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.

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

Multimodal Large Language Models (MLLMs) achieve versatility by reformulating diverse tasks into a unified instruction-following framework via instruction tuning. However, realworld deployment requires continuous adaptation to emerging tasks, motivating Multimodal Continual Instruction Tuning (MCIT). Despite its growing importance, current MCIT research is hindered by severe engineering bottlenecks. Existing methods are typically implemented by directly modifying the base MLLM codebase, which imposes substantial implementation overhead and yields methodspecific architectures that severely limit code reuse and fair comparison. To address this, we introduce PRISM, a plug-in reproducible codebase specifically designed for scalable MCIT research. It separates algorithmic development from the backbone implementation via a lightweight plugin registration mechanism, enabling new strategies to be integrated as independent plugins without modifying the underlying MLLM codebase, thereby eliminating structural fragmentation and accelerating method development. PRISM natively supports widely used large-scale training pipeline, thereby enabling reproducible and scalable MCIT experimentation. Code is available at https://github.com/LAMDA-CL/Prism.

For agents

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

get_harvested_code_for_paper("2605.26110")
get_code_for_paper("2605.26110")
have("2605.26110")

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

curl https://syntology.ai/api/ran/2605.26110.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.26110.svg)](https://syntology.ai/paper/2605.26110)

Connect an agent — have() is free.