mlsanityData to 7 Oct 2026

EcoAlign: An Economically Rational Framework for Efficient LVLM Alignment

arXiv 2511.11301 ↗First posted 14 Nov 2025cs.AIcited by 5

On the map: LLM reasoning → Reasoning for agents, planning and code

also in 2 more
Abstract (arXiv)

Large Vision-Language Models (LVLMs) exhibit powerful reasoning capabilities but suffer sophisticated jailbreak vulnerabilities. Fundamentally, aligning LVLMs is not just a safety challenge but a problem of economic efficiency. Current alignment methods struggle with the trade-off between safety, utility, and operational costs. Critically, a focus solely on final outputs (process-blindness) wastes significant computational budget on unsafe deliberation. This flaw allows harmful reasoning to be disguised with benign justifications, thereby circumventing simple additive safety scores. To address this, we propose EcoAlign, an inference-time framework that reframes alignment as an economically rational search by treating the LVLM as a boundedly rational agent. EcoAlign incrementally expands a thought graph and scores actions using a forward-looking function (analogous to net present value) that dynamically weighs expected safety, utility, and cost against the remaining budget. To prevent deception, path safety is enforced via the weakest-link principle. Extensive experiments across 3 closed-source and 2 open-source models on 6 datasets show that EcoAlign matches or surpasses state-of-the-art safety and utility at a lower computational cost, thereby offering a principled, economical pathway to robust LVLM alignment.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 46 of its 56 references that are in the map. First posted on arXiv 14 Nov 2025.

Builds on

46 of 56 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025VLMGuard-R1: Proactive Safety Alignment for VLMs via Reasoning-Driven Prompt Optimization17 Apr
  2. 2025Adversarial Reasoning at Jailbreaking Time3 Feb
  3. 2024Immune: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment27 Nov
  4. 2024Enhancing Adversarial Attacks through Chain of Thought29 Oct

Other references · 42

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2022Training language models to follow instructions with human feedbackcited by 15,712
  3. 2017Proximal Policy Optimization Algorithmscited by 14,441
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  5. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  6. 2023Visual Instruction Tuningcited by 7,792
  7. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  8. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,855
  9. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,689
  10. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,769
  11. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contextscited by 1,582
  12. 2024Are We on the Right Way for Evaluating Large Vision-Language Models?cited by 893
30 more
  1. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 873
  2. 2023Aligning Large Multimodal Models with Factually Augmented RLHFcited by 703
  3. 2023OCRBench: On the Hidden Mystery of OCR in Large Multimodal Modelscited by 466
  4. 2024LLaVA-CoT: Let Vision Language Models Reason Step-by-Stepcited by 446
  5. 2023RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedbackcited by 404
  6. 2025Chain of Draft: Thinking Faster by Writing Lesscited by 290
  7. 2024Visual Sketchpad: Sketching as a Visual Chain of Thought for Multimodal Language Modelscited by 260
  8. 2023MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Modelscited by 255
  9. 2024Mulberry: Empowering MLLM with o1-like Reasoning and Reflection via Collective Monte Carlo Tree Searchcited by 171
  10. 2024OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoningcited by 165
  11. 2023Compositional Chain-of-Thought Prompting for Large Multimodal Modelscited by 160
  12. 2024Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Modelscited by 160
  13. 2024RLAIF-V: Open-Source AI Feedback Leads to Super GPT-4V Trustworthinesscited by 129
  14. 2024Llama Guard 3 Vision: Safeguarding Human-AI Image Understanding Conversationscited by 118
  15. 2024Mind's Eye of LLMs: Visualization-of-Thought Elicits Spatial Reasoning in Large Language Modelscited by 65
  16. 2024KAM-CoT: Knowledge Augmented Multimodal Chain-of-Thoughts Reasoningcited by 60
  17. 2024ETA: Evaluating Then Aligning Safety of Vision Language Models at Inference Timecited by 57
  18. 2024LlavaGuard: An Open VLM-based Framework for Safeguarding Vision Datasets and Modelscited by 49
  19. 2025VisuoThink: Empowering LVLM Reasoning with Multimodal Tree Searchcited by 36
  20. 2024CoCA: Regaining Safety-awareness of Multimodal Large Language Models with Constitutional Calibrationcited by 30
  21. 2024Heuristic-Induced Multimodal Risk Distribution Jailbreak Attack for Multimodal Large Language Modelscited by 29
  22. 2024Unraveling and Mitigating Safety Alignment Degradation of Vision-Language Modelscited by 23
  23. 2024Reinforcement Learning from Multi-role Debates as Feedback for Bias Mitigation in LLMscited by 22
  24. 2025Inverse Reinforcement Learning with Dynamic Reward Scaling for LLM Alignmentcited by 20
  25. 2025Oyster-I: Beyond Refusal -- Constructive Safety Alignment for Responsible Language Modelscited by 19
  26. 2024PBI-Attack: Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for Toxicity Maximizationcited by 17
  27. 2025VLM-Guard: Safeguarding Vision-Language Models via Fulfilling Safety Alignment Gapcited by 16
  28. 2024AGR: Age Group fairness Reward for Bias Mitigation in LLMscited by 12
  29. 2025Strata-Sword: A Hierarchical Safety Evaluation towards LLMs based on Reasoning Complexity of Jailbreak Instructionscited by 11
  30. 2023Lever LM: Configuring In-Context Sequence to Lever Large Vision Language Modelscited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 4.

5 papers.

  1. 2026Adversarial Orthogonal Disentanglement for LVLM Hallucination Mitigationcited by 1
  2. 2026Dynamo: Dynamic Skill-Tool Evolution for Vision-Language Agentscited by 1
  3. 2025Information-Dense Reasoning for Efficient and Auditable Security Alert Triagecited by 0
  4. 2026ShellForge: Adversarial Co-Evolution of Webshell Generation and Multi-View Detection for Robust Webshell Defensecited by 0
  5. 2026Membership Inference for Contrastive Pre-training Models with Text-only PII Queriescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025VisuoAlign: Safety Alignment of LVLMs with Multimodal Tree Search
  2. 2025Think-Reflect-Revise: A Policy-Guided Reflective Framework for Safety Alignment in Large Vision Language Models
  3. 2026Two Birds, One Projection: Harmonizing Safety and Utility in LVLMs via Inference-time Feature Projection
  4. 2026Narrow Fine-Tuning Erodes Safety Alignment in Vision-Language Agents
  5. 2025MSR-Align: Policy-Grounded Multimodal Alignment for Safety-Aware Reasoning in Vision-Language Models
  6. 2024Deliberative Alignment: Reasoning Enables Safer Language Models
  7. 2025On Almost Surely Safe Alignment of Large Language Models at Inference-Time
  8. 2026Safe Context Switching for Agents in the Wild: Mitigating Subspace Interference via Orthogonal Adaptation