Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Unlocking Cognitive Capabilities and Analyzing the Perception-Logic Trade-off

Longyin Zhang, Shuo Sun, Yingxu He, Won Cheng Yi Lewis, Muhammad Huzaifah Bin Md Shahrin, Hardik Bhupendra Sailor, Heng Meng Jeremy Wong, Tarun Kumar Vangani, Yi Ma, Qiongqiong Wang, Minh Duc Pham, Ridong Jiang, Jingtao Li, Jingyi Liao, Zhuohan Liu, Yanfeng Lu, Manas Gupta, Ai Ti Aw

TL;DR

This report introduces the research preview of MERaLiON2-Omni (Alpha), a 10B-parameter multilingual omni-perception tailored for Southeast Asia, and presents a progressive training pipeline that explicitly decouples and then integrates "System 1"(Perception) and"System 2"(Reasoning) capabilities.

Abstract

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.

Unlocking Cognitive Capabilities and Analyzing the Perception-Logic Trade-off

TL;DR

This report introduces the research preview of MERaLiON2-Omni (Alpha), a 10B-parameter multilingual omni-perception tailored for Southeast Asia, and presents a progressive training pipeline that explicitly decouples and then integrates "System 1"(Perception) and"System 2"(Reasoning) capabilities.

Abstract

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.
Paper Structure (27 sections, 3 equations, 7 figures, 7 tables)

This paper contains 27 sections, 3 equations, 7 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Architecture
  3. Multilingual Language Backbone
  4. Audio Perception Module
  5. Visual Perception Module
  6. Training Methodology
  7. Phase 1: Establishing the Perception Backbone
  8. Phase 2: The Reasoning Shift
  9. Cognitive Warm-up
  10. Silver Data Alignment
  11. Data Resources & Benchmarks
  12. Perception Backbone Training Data
  13. Reasoning Instruction Set
  14. The SEA-Omni Benchmark Suite
  15. Experiments
  16. ...and 12 more sections

Figures (7)

  • Figure 1: MERaLiON2-Omni (Alpha) is a 10B omni model for Southeast Asia that extends MERaLiON-v2 with a stronger perception backbone aligning region-specific audio-visual signals to a multilingual LLM, and a Cognitive Reasoning Layer enabled by Chain-of-Thought transfer.
  • Figure 2: The cognitive gap. This radar chart illustrates the performance of our reasoning (M2O-ML-r1) and non-reasoning (M2O-ML) models on 6 representative AudioBench tasks.
  • Figure 3: The amplifier effect. The introduced reasoning capability benefits cognitive tasks but harms perceptual tasks.
  • Figure 4: Divergent impact of reasoning length. Visual accuracy improves with longer reasoning (the red line), while Audio accuracy degrades due to Temporal Drift (the green line).
  • Figure 5: The anatomy of temporal drift.
  • ...and 2 more figures