SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.15205 · NeurIPS · 2024

Leveraging Hallucinations to Reduce Manual Prompt Dependency in Promptable Segmentation

Junchi Yan, Jian Hu, Shaogang Gong, Jiayi Lin

arXiv · PDF · Open in the Atlas

Code that ran

2 of 9 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, 1 violated it and 0 have no contract verdict; 1 failure of our instrument; 7 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/2408.15205.json · badge: /api/ran/2408.15205.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
lwpyh/ProMaC_code canonical (classifier label)
classifier basis: classifier_2025
2 ran · 7 no recorded run
FunctionStatusWhere it lives
convert_patch_bbox_to_original Ran
ran, violated the contract we drafted
lwpyh/ProMaC_code/utils_mllm.py
code served (permissive licence) · get_code("923b02f12c4e8943")
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.

import torch

torch.manual_seed(0)

ARGS = {
    "bounding_box": [10, 20, 50, 80],
    "left": 5,
    "upper": 15,
}
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 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
expand_bbox Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lwpyh/ProMaC_code/utils_mllm.py
code served (permissive licence) · get_code("93aa0fead442ddef")
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 torch

torch.manual_seed(0)

# The function expand_bbox takes a bbox (list of 4 coordinates) and an expansion_rate
# It expands the bounding box by the given rate and clamps to [0, 1]

# Example bbox in normalized coordinates [x1, y1, x2, y2]
bbox_example = [0.2, 0.3, 0.7, 0.8]

# Expansion rate as a float
expansion_rate_example = 0.15

ARGS = {
    "bbox": bbox_example,
    "expansion_rate": expansion_rate_example,
}
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 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
generate_image Not yet run
unverified (harvested, no recorded run)
lwpyh/ProMaC_code/utils_mllm.py
code served (permissive licence) · get_code("4379261942974adb")
How it ran

no invocation recorded.

get_reflected_text_from_full_img_llava Not yet run
unverified (harvested, no recorded run)
lwpyh/ProMaC_code/utils_mllm.py
code served (permissive licence) · get_code("fc5b045f2c663b01")
How it ran

no invocation recorded.

get_reflected_text_from_img Not yet run
unverified (harvested, no recorded run)
lwpyh/ProMaC_code/utils_mllm.py
code served (permissive licence) · get_code("47abbc2ac21defe8")
How it ran

no invocation recorded.

get_reflected_text_from_img_llava_collected Not yet run
unverified (harvested, no recorded run)
lwpyh/ProMaC_code/utils_mllm.py
code served (permissive licence) · get_code("4f885a4ecc54c188")
How it ran

no invocation recorded.

get_text_from_img_llava_with_bbox_iter0 Not yet run
unverified (harvested, no recorded run)
lwpyh/ProMaC_code/utils_mllm.py
code served (permissive licence) · get_code("141d2009ede3317e")
How it ran

no invocation recorded.

get_text_from_img_llava_with_bbox_patch Not yet run
unverified (harvested, no recorded run)
lwpyh/ProMaC_code/utils_mllm.py
code served (permissive licence) · get_code("b470f2ff84a1db05")
How it ran

no invocation recorded.

show Not yet run
unverified (harvested, no recorded run)
lwpyh/ProMaC_code/utils_mllm.py
code served (permissive licence) · get_code("03c001cd5ce7f974")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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 9 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

Promptable segmentation typically requires instance-specific manual prompts to guide the segmentation of each desired object. To minimize such a need, taskgeneric promptable segmentation has been introduced, which employs a single task-generic prompt to segment various images of different objects in the same task. Current methods use Multimodal Large Language Models (MLLMs) to reason detailed instance-specific prompts from a task-generic prompt for improving segmentation accuracy. The effectiveness of this segmentation heavily depends on the precision of these derived prompts. However, MLLMs often suffer from hallucinations during reasoning, resulting in inaccurate prompting. While existing methods focus on eliminating hallucinations to improve a model, we argue that MLLM hallucinations can reveal valuable contextual insights when leveraged correctly, as they represent pre-trained large-scale knowledge beyond individual images. In this work, we utilize hallucinations to mine task-related information from images and verify its accuracy for enhancing precision of the generated prompts. Specifically, we introduce an iterative Prompt-Mask Cycle generation framework (ProMaC) with a prompt generator and a mask generator. The prompt generator uses a multiscale chain of thought prompting, initially exploring hallucinations for extracting extended contextual knowledge on a test image. These hallucinations are then reduced to formulate precise instance-specific prompts, directing the mask generator to produce masks that are consistent with task semantics by mask semantic alignment. The generated masks iteratively induce the prompt generator to focus more on task-relevant image areas and reduce irrelevant hallucinations, jointly resulting in better prompts and masks. Experiments on 5 benchmarks demonstrate the effectiveness of ProMaC. Code given in https://lwpyh.github.io/ProMaC/.

For agents

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

get_harvested_code_for_paper("2408.15205")
get_code_for_paper("2408.15205")
have("2408.15205")

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

curl https://syntology.ai/api/ran/2408.15205.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.15205.svg)](https://syntology.ai/paper/2408.15205)

Connect an agent — have() is free.