mlsanityData to 14 Oct 2026

Order Is Not Control: Driven-Dissipative Response Laws Across Artificial and Biological Systems

arXiv 2606.12923 ↗First posted 11 Jun 2026cs.LG cs.AI cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

AI alignment, interpretability, steering, and neural perturbation studies identify order-inducing objects. We argue that order is not control. Control requires a receiver-gated response law: a denominator-indexed operator mapping material state, action/drive, bath, and receiver state to response displacement, sinks, effort, and basin projection. We identify it across biological, LLM, adapter, and stochastic-operator panels. The laws are local: an intervention can be admitted, saturated, sign-changing, leaky, or overdriven depending on medium, bath, receiver state, action port, and comparator. Control is assigned when finite effort moves a target or outcome-readout class under the same denominator while damage, null/evasive, invalid format, overdrive, and unnecessary effort stay bounded. Mouse ALM, C. elegans, and zebrafish panels provide physical response-operator evidence while excluding coordinate identity and controller conclusions. LLM panels show generated-output response laws: across four material conditions, response vectors are predictable at 72.8-73.7% component-sign accuracy, rising to 84.3-84.8% on nonzero components; held-out observers predict system-effect and target/oracle families at 93.6% and 91.7% accuracy. Constitution-conditioned adapters reshape susceptibility as prepared media, and stochastic-operator panels separate measured opportunity from deployable action policies. This gives a driven-dissipative response-system account at the mesoscopic control level: drives act through prepared media, baths, and receivers, producing admitted movement, impedance, sinks, or overdrive. The evidence supports local admitted control and measurable stochastic response operators, while leaving deployable pre-generation control, hidden/logit causal sufficiency, biological-to-LLM coordinate identity, and literal thermodynamic quantities outside scope.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 32 of its 74 references that are in the map. First posted on arXiv 11 Jun 2026.

Builds on

32 of 74 references.

References · 32

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  3. 2025Qwen3 Technical Reportcited by 8,820
  4. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  5. 2023Let's Verify Step by Stepcited by 3,789
  6. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
  7. 2025Gemma 3 Technical Reportcited by 1,824
  8. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,336
  9. 2024Refusal in Language Models Is Mediated by a Single Directioncited by 1,032
  10. 2023Steering Language Models With Activation Engineeringcited by 922
  11. 2025Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAscited by 521
  12. 2024Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agentscited by 369
20 more
  1. 2024Alignment faking in large language modelscited by 360
  2. 2023Towards Best Practices of Activation Patching in Language Models: Metrics and Methodscited by 359
  3. 2024Programming Refusal with Conditional Activation Steeringcited by 206
  4. 2024Collective Constitutional AI: Aligning a Language Model with Public Inputcited by 123
  5. 2025SycEval: Evaluating LLM Sycophancycited by 108
  6. 2025Measuring Sycophancy of Language Models in Multi-turn Dialoguescited by 104
  7. 2025Thought Crime: Backdoors and Emergent Misalignment in Reasoning Modelscited by 65
  8. 2025FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoningcited by 58
  9. 2025Harnessing the Universal Geometry of Embeddingscited by 55
  10. 2026Revisiting the Platonic Representation Hypothesis: An Aristotelian Viewcited by 44
  11. 2025Steering off Course: Reliability Challenges in Steering Language Modelscited by 34
  12. 2024Learning stochastic dynamics from snapshots through regularized unbalanced optimal transportcited by 33
  13. 2025Why Do Some Language Models Fake Alignment While Others Don't?cited by 20
  14. 2024Self-Supervised Alignment with Mutual Information: Learning to Follow Principles without Preference Labelscited by 19
  15. 2025DeceptionBench: A Comprehensive Benchmark for AI Deception Behaviors in Real-world Scenarioscited by 19
  16. 2026Measuring AI R&D Automationcited by 9
  17. 2024ABC Align: Large Language Model Alignment for Safety & Accuracycited by 4
  18. 2025ENIGMA: The Geometry of Reasoning and Alignment in Large-Language Modelscited by 2
  19. 2026Inspectable AI for Science: A Research Object Approach to Generative AI Governancecited by 2
  20. 2026ATLAS: Constitution-Conditioned Latent Geometry and Redistribution Across Language Models and Neural Perturbation Datacited by 1

Built on by

Not cited yet.