% Key papers for Mechanistic analysis of attention and representations (scime atlas, https://atlas.scime.ai). Authors are not included: add them from the paper pages.

@inproceedings{neurips2019_anti_57900,
  title = {{Anti-efficient encoding in emergent communication}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2019},
  url = {https://nips.cc/virtual/2019/poster/13743},
}

@inproceedings{icml2026_mechanistic_51741,
  title = {{Mechanistic Data Attribution: Tracing the Training Origins of Interpretable LLM Units}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=PQaxfoEcRc},
}

@inproceedings{neurips2024_unibias_73703,
  title = {{UniBias: Unveiling and Mitigating LLM Bias through Internal Attention and FFN Manipulation}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/93794},
}

@inproceedings{neurips2025_why_76504,
  title = {{Why and How LLMs Hallucinate: Connecting the Dots with Subsequence Associations}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=K2Q4Jp4RbB},
}

@inproceedings{neurips2024_adversarial_73088,
  title = {{Adversarial Representation Engineering: A General Model Editing Framework for Large Language Models}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/94340},
}

@inproceedings{neurips2025_concept_79312,
  title = {{Concept-Guided Interpretability via Neural Chunking}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=o87dDXYLXC},
}

@inproceedings{icml2025_towards_48067,
  title = {{Towards scientific discovery with dictionary learning: Extracting biological concepts from microscopy foundation models}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/44540},
}

@inproceedings{neurips2025_cognitive_75974,
  title = {{Cognitive Mirrors: Exploring the Diverse Functional Roles of Attention Heads in LLM Reasoning}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=EBwfFrw5VA},
}

@inproceedings{neurips2025_who_77757,
  title = {{Who Reasons in the Large Language Models?}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=XIqlxqNDCL},
}

@inproceedings{neurips2025_language_79545,
  title = {{Language Models Are Capable of Metacognitive Monitoring and Control of Their Internal Activations}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=qTXlFwlggv},
}

@inproceedings{iclr2025_programming_81979,
  title = {{Programming Refusal with Conditional Activation Steering}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2025},
  url = {https://iclr.cc/virtual/2025/poster/29806},
}

@inproceedings{aaai2025_benchmarking_8242,
  title = {{Benchmarking and Understanding Compositional Relational Reasoning of LLMs}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2025},
  url = {https://ojs.aaai.org/index.php/AAAI/article/view/34170},
}

@inproceedings{aaai2025_on_10166,
  title = {{On Effects of Steering Latent Representation for Large Language Model Unlearning}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2025},
  url = {https://ojs.aaai.org/index.php/AAAI/article/view/34544},
}

@inproceedings{icml2025_eliciting_46420,
  title = {{Eliciting Language Model Behaviors with Investigator Agents}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/46145},
}

@inproceedings{icml2025_jacobian_47418,
  title = {{Jacobian Sparse Autoencoders: Sparsify Computations, Not Just Activations}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/45175},
}

@inproceedings{icml2026_beyond_49969,
  title = {{Beyond Scalars: Evaluating and Understanding LLM Reasoning via Geometric Progress and Stability}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=7i44Lx4KEY},
}

@inproceedings{icml2026_sparse_50972,
  title = {{Sparse Autoencoders for Interpretable Emotion Control in Text-to-Speech}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=Hbt2ryJgz2},
}

@inproceedings{icml2026_efficient_51078,
  title = {{Efficient LLM Moderation with Multi-Layer Latent Prototypes}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=IdfwoUzREG},
}

@inproceedings{icml2026_dual_51094,
  title = {{Dual Mechanisms of Value Expression: Intrinsic vs. Prompted Values in Large Language Models}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=IjLOXSP41n},
}

@inproceedings{icml2026_foeglass_51149,
  title = {{FoeGlass: Simple In-Context Learning Is Enough for Red Teaming Audio Deepfake Detectors}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=J6amahDTKV},
}

@inproceedings{icml2026_prompt_51396,
  title = {{Prompt Injection as Role Confusion}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=Lm3DQDyIsI},
}