Divide and Conquer: Exploring Language-centric Tree Reasoning for Video Question-Answering

Zhaohe Liao, Jiangtong Li, Siyu Sun, Qingyang Liu, Fengshun Xiao, Tianjiao Li, Qiang Zhang, Guang Chen, Li Niu, Changjun Jiang, Liqing Zhang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:37511-37527, 2025.

Abstract

Video Question-Answering (VideoQA) remains challenging in achieving advanced cognitive reasoning due to the uncontrollable and opaque reasoning processes in existing Multimodal Large Language Models (MLLMs). To address this issue, we propose a novel Language-centric Tree Reasoning (LTR) framework that targets on enhancing the reasoning ability of models. In detail, it recursively divides the original question into logically manageable parts and conquers them piece by piece, enhancing the reasoning capabilities and interpretability of existing MLLMs. Specifically, in the first stage, the LTR focuses on language to recursively generate a language-centric logical tree, which gradually breaks down the complex cognitive question into simple perceptual ones and plans the reasoning path through a RAG-based few-shot approach. In the second stage, with the aid of video content, the LTR performs bottom-up logical reasoning within the tree to derive the final answer along with the traceable reasoning path. Experiments across 11 VideoQA benchmarks demonstrate that our LTR framework significantly improves both accuracy and interpretability compared to state-of-the-art MLLMs. To our knowledge, this is the first work to implement a language-centric logical tree to guide MLLM reasoning in VideoQA, paving the way for language-centric video understanding from perception to cognition.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-liao25d, title = {Divide and Conquer: Exploring Language-centric Tree Reasoning for Video Question-Answering}, author = {Liao, Zhaohe and Li, Jiangtong and Sun, Siyu and Liu, Qingyang and Xiao, Fengshun and Li, Tianjiao and Zhang, Qiang and Chen, Guang and Niu, Li and Jiang, Changjun and Zhang, Liqing}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {37511--37527}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/liao25d/liao25d.pdf}, url = {https://proceedings.mlr.press/v267/liao25d.html}, abstract = {Video Question-Answering (VideoQA) remains challenging in achieving advanced cognitive reasoning due to the uncontrollable and opaque reasoning processes in existing Multimodal Large Language Models (MLLMs). To address this issue, we propose a novel Language-centric Tree Reasoning (LTR) framework that targets on enhancing the reasoning ability of models. In detail, it recursively divides the original question into logically manageable parts and conquers them piece by piece, enhancing the reasoning capabilities and interpretability of existing MLLMs. Specifically, in the first stage, the LTR focuses on language to recursively generate a language-centric logical tree, which gradually breaks down the complex cognitive question into simple perceptual ones and plans the reasoning path through a RAG-based few-shot approach. In the second stage, with the aid of video content, the LTR performs bottom-up logical reasoning within the tree to derive the final answer along with the traceable reasoning path. Experiments across 11 VideoQA benchmarks demonstrate that our LTR framework significantly improves both accuracy and interpretability compared to state-of-the-art MLLMs. To our knowledge, this is the first work to implement a language-centric logical tree to guide MLLM reasoning in VideoQA, paving the way for language-centric video understanding from perception to cognition.} }
Endnote
%0 Conference Paper %T Divide and Conquer: Exploring Language-centric Tree Reasoning for Video Question-Answering %A Zhaohe Liao %A Jiangtong Li %A Siyu Sun %A Qingyang Liu %A Fengshun Xiao %A Tianjiao Li %A Qiang Zhang %A Guang Chen %A Li Niu %A Changjun Jiang %A Liqing Zhang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-liao25d %I PMLR %P 37511--37527 %U https://proceedings.mlr.press/v267/liao25d.html %V 267 %X Video Question-Answering (VideoQA) remains challenging in achieving advanced cognitive reasoning due to the uncontrollable and opaque reasoning processes in existing Multimodal Large Language Models (MLLMs). To address this issue, we propose a novel Language-centric Tree Reasoning (LTR) framework that targets on enhancing the reasoning ability of models. In detail, it recursively divides the original question into logically manageable parts and conquers them piece by piece, enhancing the reasoning capabilities and interpretability of existing MLLMs. Specifically, in the first stage, the LTR focuses on language to recursively generate a language-centric logical tree, which gradually breaks down the complex cognitive question into simple perceptual ones and plans the reasoning path through a RAG-based few-shot approach. In the second stage, with the aid of video content, the LTR performs bottom-up logical reasoning within the tree to derive the final answer along with the traceable reasoning path. Experiments across 11 VideoQA benchmarks demonstrate that our LTR framework significantly improves both accuracy and interpretability compared to state-of-the-art MLLMs. To our knowledge, this is the first work to implement a language-centric logical tree to guide MLLM reasoning in VideoQA, paving the way for language-centric video understanding from perception to cognition.
APA
Liao, Z., Li, J., Sun, S., Liu, Q., Xiao, F., Li, T., Zhang, Q., Chen, G., Niu, L., Jiang, C. & Zhang, L.. (2025). Divide and Conquer: Exploring Language-centric Tree Reasoning for Video Question-Answering. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:37511-37527 Available from https://proceedings.mlr.press/v267/liao25d.html.

Related Material