mlsanityData to 7 Oct 2026

Cognitive Load Traces as Symbolic and Visual Accounts of Deep Model Cognition

arXiv 2510.15980 ↗First posted 13 Oct 2025cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

We propose Cognitive Load Traces (CLTs) as a mid-level interpretability framework for deep models, inspired by Cognitive Load Theory in human cognition. CLTs are defined as symbolic, temporally varying functions that quantify model-internal resource allocation. Formally, we represent CLTs as a three-component stochastic process (ILt, ELt, GLt), corresponding to Intrinsic, Extraneous, and Germane load. Each component is instantiated through measurable proxies such as attention entropy, KV-cache miss ratio, representation dispersion, and decoding stability. We propose both symbolic formulations and visualization methods (load curves, simplex diagrams) that enable interpretable analysis of reasoning dynamics. Experiments on reasoning and planning benchmarks show that CLTs predict error-onset, reveal cognitive strategies, and enable load-guided interventions that improve reasoning efficiency by 15-30% while maintaining accuracy.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 26 of its 35 references that are in the map. First posted on arXiv 13 Oct 2025.

Builds on

26 of 35 references.

References · 26

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2023GPT-4 Technical Reportcited by 18,132
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  5. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  6. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  7. 2024Mixtral of Expertscited by 1,785
  8. 2023PaLM 2 Technical Reportcited by 1,154
  9. 2019Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Prunedcited by 844
  10. 2019Are Sixteen Heads Really Better than One?cited by 803
  11. 2025FastCache: Fast Caching for Diffusion Transformer Through Learnable Linear Approximationcited by 23
  12. 2024Advances in APPFL: A Comprehensive and Extensible Federated Learning Frameworkcited by 19
14 more
  1. 2024MT2ST: Adaptive Multi-Task to Single-Task Learningcited by 16
  2. 2024LLMEasyQuant: Scalable Quantization for Parallel and Distributed LLM Inferencecited by 15
  3. 2024HADES: Hardware Accelerated Decoding for Efficient Speculation in Large Language Modelscited by 15
  4. 2025Breaking Silos: Adaptive Model Fusion Unlocks Better Time Series Forecastingcited by 12
  5. 2025Scalability Optimization in Cloud-Based AI Inference Services: Strategies for Real-Time Load Balancing and Automated Scalingcited by 10
  6. 2025Cloud-Based AI Systems: Leveraging Large Language Models for Intelligent Fault Detection and Autonomous Self-Healingcited by 10
  7. 2025Cross-Cloud Data Privacy Protection: Optimizing Collaborative Mechanisms of AI Systems by Integrating Federated Learning and LLMscited by 9
  8. 2025TinyServe: Query-Aware Cache Selection for Efficient LLM Servingcited by 9
  9. 2024Designing Large Foundation Models for Efficient Training and Inference: A Surveycited by 7
  10. 2025PiKV: KV Cache Management System for Mixture of Expertscited by 5
  11. 2025HSGM: Hierarchical Segment-Graph Memory for Scalable Long-Text Semanticscited by 5
  12. 2024GraphSnapShot: Caching Local Structure for Fast Graph Learningcited by 4
  13. 2025SemToken: Semantic-Aware Tokenization for Efficient Long-Context Language Modelingcited by 3
  14. 2025QuickMerge++: Fast Token Merging with Autoregressive Priorcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Cognitive Load-Aware Inference: A Neuro-Symbolic Framework for Optimizing the Token Economy of Large Language Models
  2. 2025CogniLoad: A Synthetic Natural Language Reasoning Benchmark With Tunable Length, Intrinsic Difficulty, and Distractor Density
  3. 2025Cognitive Load Limits in Large Language Models: Benchmarking Multi-Hop Reasoning
  4. 2025United Minds or Isolated Agents? Exploring Coordination of LLMs under Cognitive Load Theory
  5. 2026Monitoring the Internal Monologue: Probe Trajectories Reveal Reasoning Dynamics
  6. 2026Beyond Accuracy: A Cognitive Load Framework for Mapping the Capability Boundaries of Tool-use Agents
  7. 2025Visual Categorization Across Minds and Models: Cognitive Analysis of Human Labeling and Neuro-Symbolic Integration
  8. 2026From Decorative to Load-Bearing: Task Difficulty Shapes the Causal Role of Chain-of-Thought