Chapter 40: Further Reading

Test-Time Compute and Inference Scaling

Snell, C., Lee, J., Xu, K., & Kumar, A. (2024). "Scaling LLM Test-Time Compute Optimally Can Be More Effective Than Scaling Model Parameters." arXiv preprint arXiv:2408.03314. A rigorous analysis demonstrating that allocating compute at inference can be more cost-effective than training larger models.
Wei, J., Wang, X., Schuurmans, D., et al. (2022). "Chain-of-Thought Prompting Elicits Reasoning in Large Language Models." NeurIPS 2022. The foundational paper on chain-of-thought prompting, showing that intermediate reasoning steps improve performance on multi-step problems.
Brown, B., Juravsky, J., Ehrlich, R., et al. (2024). "Large Language Monkeys: Scaling Inference Compute with Repeated Sampling." arXiv preprint arXiv:2407.21787. Demonstrates that repeated sampling with verification can dramatically improve performance on coding and math benchmarks.

Ha, D. & Schmidhuber, J. (2018). "World Models." arXiv preprint arXiv:1803.10122. A seminal paper introducing the modern concept of learned world models for reinforcement learning agents.
Bruce, J., Dennis, M., Edwards, A., et al. (2024). "Genie: Generative Interactive Environments." arXiv preprint arXiv:2402.15391. Demonstrates learning interactive world models from unlabeled video data.
NVIDIA. (2024). "Cosmos: World Foundation Models." Technical Report. NVIDIA's approach to building world foundation models for physical AI and robotics.

Garcez, A. d'A. & Lamb, L. C. (2023). "Neurosymbolic AI: The 3rd Wave." Artificial Intelligence Review, 56, 12387--12406. A comprehensive survey of neurosymbolic integration approaches and their historical context.
Li, Z., Huang, J., & Naik, M. (2023). "Scallop: A Language for Neurosymbolic Programming." PLDI 2023. Introduces Scallop, a probabilistic programming language for neurosymbolic applications.
Gao, L., Madaan, A., Zhou, S., et al. (2023). "PAL: Program-Aided Language Models." ICML 2023. Demonstrates LLMs generating code to offload computation to symbolic executors.

Kirkpatrick, J., Pascanu, R., Rabinowitz, N., et al. (2017). "Overcoming Catastrophic Forgetting in Neural Networks." PNAS, 114(13), 3521--3526. The original EWC paper, introducing Fisher information regularization for continual learning.
De Lange, M., Aljundi, R., Masana, M., et al. (2022). "A Continual Learning Survey: Defying Forgetting in Classification Tasks." IEEE TPAMI, 44(7), 3366--3385. A comprehensive survey covering regularization, replay, and architecture-based continual learning methods.
Wang, L., Zhang, X., Su, H., & Zhu, J. (2024). "A Comprehensive Survey of Continual Learning: Theory, Method and Application." IEEE TPAMI. An updated survey covering continual learning in the era of foundation models.

Jumper, J., Evans, R., Pritzel, A., et al. (2021). "Highly Accurate Protein Structure Prediction with AlphaFold." Nature, 596, 583--589. The paper that solved the protein folding problem using deep learning.
Merchant, A., Batzner, S., Schoenholz, S. S., et al. (2023). "Scaling Deep Learning for Materials Discovery." Nature, 624, 80--85. GNoME's discovery of millions of new stable materials using graph neural networks.
Lu, C., Lu, C., Lange, R. T., et al. (2024). "The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery." arXiv preprint arXiv:2408.06292. An early prototype of an autonomous AI system for scientific research.

Yao, S., Zhao, J., Yu, D., et al. (2023). "ReAct: Synergizing Reasoning and Acting in Language Models." ICLR 2023. A foundational framework for language model agents that interleave reasoning and tool use.
Jimenez, C. E., Yang, J., Wettig, A., et al. (2024). "SWE-bench: Can Language Models Resolve Real-World GitHub Issues?" ICLR 2024. A benchmark for evaluating coding agents on real software engineering tasks.
Zhou, S., Xu, F. F., Zhu, H., et al. (2024). "WebArena: A Realistic Web Environment for Building Autonomous Agents." ICLR 2024. A benchmark for evaluating web-browsing AI agents.

Morris, M. R., Sohl-Dickstein, J., Fiedel, N., et al. (2024). "Levels of AGI: Operationalizing Progress on the Path to AGI." arXiv preprint arXiv:2311.02462. A framework for measuring AGI progress using a graduated capability scale.
Sutskever, I. (2024). "An Observation on Generalization." Various talks and interviews on the scaling hypothesis and its limits.
Chollet, F. (2019). "On the Measure of Intelligence." arXiv preprint arXiv:1911.01547. A thoughtful analysis of what intelligence measurement should capture, beyond narrow benchmarks.

Cerezo, M., Arrasmith, A., Babbush, R., et al. (2021). "Variational Quantum Algorithms." Nature Reviews Physics, 3, 625--644. A comprehensive review of variational quantum algorithms including quantum ML approaches.
Schuld, M. & Petruccione, F. (2021). Machine Learning with Quantum Computers, 2nd ed. Springer. A textbook covering quantum ML from foundations to applications.
McClean, J. R., Boixo, S., Smelyanskiy, V. N., et al. (2018). "Barren Plateaus in Quantum Neural Network Training Landscapes." Nature Communications, 9, 4812. The paper identifying the barren plateau problem in parameterized quantum circuits.

Ng, A. (2024). "How to Build Your Career in AI." DeepLearning.AI Reading List. Practical advice on building an AI career from one of the field's most influential educators.
Huyen, C. (2022). Designing Machine Learning Systems. O'Reilly Media. An excellent guide to the practical challenges of building production ML systems.

European Union. (2024). "EU Artificial Intelligence Act." Official regulation text. The first comprehensive AI regulation, establishing risk-based requirements for AI systems.
Weidinger, L., Mellor, J., Rauh, M., et al. (2021). "Ethical and Social Risks of Harm from Language Models." arXiv preprint arXiv:2112.04359. A taxonomy of risks from large language models with mitigation strategies.
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). "On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?" FAccT 2021. An influential paper on the environmental and social costs of large language models.