Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.23730 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914356312145920 |
|---|---|
| author | Zhang, Longyin Sun, Shuo He, Yingxu Lewis, Won Cheng Yi Shahrin, Muhammad Huzaifah Bin Md Sailor, Hardik Bhupendra Wong, Heng Meng Jeremy Vangani, Tarun Kumar Ma, Yi Wang, Qiongqiong Pham, Minh Duc Jiang, Ridong Li, Jingtao Liao, Jingyi Liu, Zhuohan Lu, Yanfeng Gupta, Manas Aw, Ai Ti |
| author_facet | Zhang, Longyin Sun, Shuo He, Yingxu Lewis, Won Cheng Yi Shahrin, Muhammad Huzaifah Bin Md Sailor, Hardik Bhupendra Wong, Heng Meng Jeremy Vangani, Tarun Kumar Ma, Yi Wang, Qiongqiong Pham, Minh Duc Jiang, Ridong Li, Jingtao Liao, Jingyi Liu, Zhuohan Lu, Yanfeng Gupta, Manas Aw, Ai Ti |
| contents | Recent advancements in Multimodal Large Language Models (MLLMs) pursue omni-perception capabilities, yet integrating robust sensory grounding with complex reasoning remains a challenge, particularly for underrepresented regions. In this report, we introduce the research preview of MERaLiON2-Omni (Alpha), a 10B-parameter multilingual omni-perception tailored for Southeast Asia (SEA). We present a progressive training pipeline that explicitly decouples and then integrates "System 1" (Perception) and "System 2" (Reasoning) capabilities. First, we establish a robust Perception Backbone by aligning region-specific audio-visual cues (e.g., Singlish code-switching, local cultural landmarks) with a multilingual LLM through orthogonal modality adaptation. Second, to inject cognitive capabilities without large-scale supervision, we propose a cost-effective Generate-Judge-Refine pipeline. By utilizing a Super-LLM to filter hallucinations and resolve conflicts via a consensus mechanism, we synthesize high-quality silver data that transfers textual Chain-of-Thought reasoning to multimodal scenarios.
Comprehensive evaluation on our newly introduced SEA-Omni Benchmark Suite reveals an Efficiency-Stability Paradox: while reasoning acts as a non-linear amplifier for abstract tasks (boosting mathematical and instruction-following performance significantly), it introduces instability in low-level sensory processing. Specifically, we identify Temporal Drift in long-context audio, where extended reasoning desynchronizes the model from acoustic timestamps, and Visual Over-interpretation, where logic overrides pixel-level reality. This report details the architecture, the data-efficient training recipe, and a diagnostic analysis of the trade-offs between robust perception and structured reasoning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_23730 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Unlocking Cognitive Capabilities and Analyzing the Perception-Logic Trade-off Zhang, Longyin Sun, Shuo He, Yingxu Lewis, Won Cheng Yi Shahrin, Muhammad Huzaifah Bin Md Sailor, Hardik Bhupendra Wong, Heng Meng Jeremy Vangani, Tarun Kumar Ma, Yi Wang, Qiongqiong Pham, Minh Duc Jiang, Ridong Li, Jingtao Liao, Jingyi Liu, Zhuohan Lu, Yanfeng Gupta, Manas Aw, Ai Ti Artificial Intelligence Recent advancements in Multimodal Large Language Models (MLLMs) pursue omni-perception capabilities, yet integrating robust sensory grounding with complex reasoning remains a challenge, particularly for underrepresented regions. In this report, we introduce the research preview of MERaLiON2-Omni (Alpha), a 10B-parameter multilingual omni-perception tailored for Southeast Asia (SEA). We present a progressive training pipeline that explicitly decouples and then integrates "System 1" (Perception) and "System 2" (Reasoning) capabilities. First, we establish a robust Perception Backbone by aligning region-specific audio-visual cues (e.g., Singlish code-switching, local cultural landmarks) with a multilingual LLM through orthogonal modality adaptation. Second, to inject cognitive capabilities without large-scale supervision, we propose a cost-effective Generate-Judge-Refine pipeline. By utilizing a Super-LLM to filter hallucinations and resolve conflicts via a consensus mechanism, we synthesize high-quality silver data that transfers textual Chain-of-Thought reasoning to multimodal scenarios. Comprehensive evaluation on our newly introduced SEA-Omni Benchmark Suite reveals an Efficiency-Stability Paradox: while reasoning acts as a non-linear amplifier for abstract tasks (boosting mathematical and instruction-following performance significantly), it introduces instability in low-level sensory processing. Specifically, we identify Temporal Drift in long-context audio, where extended reasoning desynchronizes the model from acoustic timestamps, and Visual Over-interpretation, where logic overrides pixel-level reality. This report details the architecture, the data-efficient training recipe, and a diagnostic analysis of the trade-offs between robust perception and structured reasoning. |
| title | Unlocking Cognitive Capabilities and Analyzing the Perception-Logic Trade-off |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2602.23730 |