We lifted 9 functions out of this paper's own repositories and ran 5 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| AceCHQ/MMIQ | canonical | 5 of 9 |
| Function | Status | Where it lives |
|---|---|---|
| extract_choice | Ran | AceCHQ/MMIQ/mmiq/utils/eval_utils.py code served (permissive licence) · get_code("98c18be2c0a33db2") |
| llava_image_processor | Ran | AceCHQ/MMIQ/mmiq/utils/model_utils.py code served (permissive licence) · get_code("e4e30d1813045cfb") |
| parse_multi_choice_response | Ran | AceCHQ/MMIQ/mmiq/utils/eval_utils.py code served (permissive licence) · get_code("c2cf50db9a54f717") |
| process_single_sample | Ran | AceCHQ/MMIQ/mmiq/utils/data_utils.py code served (permissive licence) · get_code("09fe27c6ee70aa9b") |
| run_model | Ran | AceCHQ/MMIQ/mmiq/run_llava.py code served (permissive licence) · get_code("fdb3e1f2061eee8e") |
| call_llava_engine_df | Not yet run | AceCHQ/MMIQ/mmiq/utils/model_utils.py code served (permissive licence) · get_code("7712b2cda4dfffde") |
| check_is_number | Not yet run | AceCHQ/MMIQ/mmiq/utils/eval_utils.py code served (permissive licence) · get_code("81305e2c0d5ef1a9") |
| load_yaml | Not yet run | AceCHQ/MMIQ/mmiq/utils/data_utils.py code served (permissive licence) · get_code("ce93b7e192e1a8b7") |
| parse_img_path | Not yet run | AceCHQ/MMIQ/mmiq/utils/data_utils.py code served (permissive licence) · get_code("de4427887bed9253") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
IQ testing has served as a foundational methodology for evaluating human cognitive capabilities, deliberately decoupling assessment from linguistic background, language proficiency, or domain-specific knowledge to isolate core competencies in abstraction and reasoning. Yet, artificial intelligence research currently lacks systematic benchmarks to quantify these critical cognitive capabilities in multimodal systems. To address this crucial gap, we propose MM-IQ, a comprehensive evaluation framework, which comprises a large-scale training set with 4,776 visual reasoning problems and 2,710 meticulously curated test items spanning 8 distinct reasoning paradigms. Through systematic evaluation of existing open-source and proprietary multimodal models, our benchmark reveals striking limitations: even state-of-the-art architectures achieve only marginally superior performance to random chance (33.17% vs. 25% baseline accuracy). This substantial performance chasm highlights the inadequacy of current multimodal models in approximating fundamental human reasoning capacities, underscoring the need for paradigm-shifting advancements to bridge this cognitive divide. Moreover, inspired by the recent surge of large reasoning models, we also release a multimodal reasoning model as the baseline that is trained via reinforcement learning with verifiable reward functions, reaching competitive performance to the state-of-the-art with a notably smaller model size.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2502.00698")
get_code_for_paper("2502.00698")
have("2502.00698")
Connect an agent — have() is free.