corpus.blog
41554 blogs · [ { "id": "01a0cfc8-83ee-70e5-8c70-9a071f38051b", "title": "Graphical Bayesian Networks for Predicting Asset Covariances", "url": "https://fanpu.io/projects/bayesian_networks_covariance_prediction_project/", "published_at": "2026-09-19T13:34:14+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07203686d1", "title": "Improving Domain Adaptation of Transformer Models For Generating Reddit Comments", "url": "https://fanpu.io/projects/improving_domain_adaptation_of_transformer_models_for_generating_reddit_commentsproject/", "published_at": "2026-09-19T13:34:14+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0720fcee12", "title": "Pseudo-determinism for Graph Streaming Problems", "url": "https://fanpu.io/projects/pseudodeterminism_for_graph_streaming_problems/", "published_at": "2026-09-19T13:34:14+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07216d9ea9", "title": "Analysis of Symmetry and Conventions in Off-Belief Learning in Hanabi", "url": "https://fanpu.io/projects/symmetry_and_conventions_in_obl_in_hanabi_project/", "published_at": "2026-09-19T13:34:14+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07598bb6fa", "title": "Blackjack trainer", "url": "https://fanpu.io/games/blackjack/", "published_at": "2026-09-19T13:33:47+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a075a5aa8ed", "title": "Fermi Estimates", "url": "https://fanpu.io/games/fermi/", "published_at": "2026-09-19T13:33:47+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a075afba2b0", "title": null, "url": "https://fanpu.io/games/flyjack/", "published_at": "2026-09-19T13:33:47+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a075bafc1f5", "title": "Parallelism analyzer", "url": "https://fanpu.io/games/parallelism/", "published_at": "2026-09-19T13:33:47+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a075c7ded74", "title": "Poker Trainer", "url": "https://fanpu.io/games/poker/", "published_at": "2026-09-19T13:33:47+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a075d311fdd", "title": "流觞 · liú shāng", "url": "https://fanpu.io/games/the-floating-cup/", "published_at": "2026-09-19T13:33:47+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a075dd83217", "title": "Xiangqi trainer", "url": "https://fanpu.io/games/xiangqi/", "published_at": "2026-09-19T13:33:47+00:00" }, { "id": "01a0c50c-6fd5-72dc-a906-b87a8ee14846", "title": "Research Sprint Day 4 - ???", "url": "https://fanpu.io/blog/2026/research-day4-gdn-baselines/", "published_at": "2026-09-18T00:00:00+00:00" }, { "id": "01a0c50c-6fd5-72dc-a906-b87a8ef2b3f6", "title": "Research Sprint Day 3 - Seed Noise and Transformer Baselines", "url": "https://fanpu.io/blog/2026/research-day3-seed-noise-and-transformer-baselines/", "published_at": "2026-09-12T00:00:00+00:00" }, { "id": "01a09630-9df4-722c-803c-8ef39a89bfa2", "title": "Research Sprint Day 2 - Rooflines", "url": "https://fanpu.io/blog/2026/research-day2-rooflines/", "published_at": "2026-09-11T00:00:00+00:00" }, { "id": "01a08c9e-400c-7184-b8dd-85f7bf94d4e3", "title": "Day 1 - Verifying the flash-linear-attention Gated DeltaNet kernel on GB10 (DGX Spark)", "url": "https://fanpu.io/blog/2026/research-day1-gdn-kernel-numerics/", "published_at": "2026-09-10T00:00:00+00:00" }, { "id": "01a08c9e-400c-7184-b8dd-85f7bf959eb1", "title": "Concise Introduction to Gated DeltaNets", "url": "https://fanpu.io/blog/2026/gated-deltanet/", "published_at": "2026-09-09T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07416fe527", "title": "Diffusion-LM Improves Controllable Text Generation", "url": "https://fanpu.io/summaries/2025-06-08-diffusion-lm-improves-controllable-text-generation/", "published_at": "2025-06-08T00:00:00+00:00" }, { "id": "01a08776-ea0a-718b-9349-da21ceeb4c2e", "title": "Neural Networks from Maximizing Rate Reduction", "url": "https://fanpu.io/blog/2025/neural-networks-by-maximizing-rate-reduction/", "published_at": "2025-05-10T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0702e3ffa4", "title": "Redunet Post | Fan Pu Zeng", "url": "https://fanpu.io/news/redunet-post/", "published_at": "2025-05-10T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07413a1e73", "title": "ReduNet: A White-box Deep Network from the Principle of Maximizing Rate Reduction", "url": "https://fanpu.io/summaries/2025-05-10-redunet-a-white-box-deep-network-from-the-principle-of-maximizing-rate-reduction/", "published_at": "2025-05-10T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0740f5ff88", "title": "DiLoCo: Distributed Low-Communication Training of Language Models", "url": "https://fanpu.io/summaries/2025-05-05-diloco-distributed-low-communication-training-of-language-models/", "published_at": "2025-05-05T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0702584ca5", "title": "Iclr 2025 | Fan Pu Zeng", "url": "https://fanpu.io/news/iclr-2025/", "published_at": "2025-04-20T00:00:00+00:00" }, { "id": "01a08776-ea0a-718b-9349-da21cf8ba3a5", "title": "An Intuitive Introduction to Gaussian Processes", "url": "https://fanpu.io/blog/2025/gaussian-processes/", "published_at": "2025-01-21T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07019b3dd0", "title": "Gaussian Process | Fan Pu Zeng", "url": "https://fanpu.io/news/gaussian-process/", "published_at": "2025-01-21T00:00:00+00:00" }, { "id": "01a08776-ea0a-718b-9349-da21d004f049", "title": "Bounding Mixing Times of Markov Chains via the Spectral Gap", "url": "https://fanpu.io/blog/2025/bounding-markov-chain-mixing-times-by-spectral-gap/", "published_at": "2025-01-12T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07010eccf8", "title": "Markov Chain Mixing Times | Fan Pu Zeng", "url": "https://fanpu.io/news/markov-chain-mixing-times/", "published_at": "2025-01-12T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07006d924e", "title": "Neurips 2024 | Fan Pu Zeng", "url": "https://fanpu.io/news/neurips-2024/", "published_at": "2024-11-29T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07401b01bd", "title": "RAGAS: Automated Evaluation of Retrieval Augmented Generation", "url": "https://fanpu.io/summaries/2024-11-04-ragas-automated-evaluation-of-retrieval-augmented-generation/", "published_at": "2024-11-04T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073f2bec4b", "title": "Easy-to-Hard Generalization: Scalable Alignment Beyond Human Supervision", "url": "https://fanpu.io/summaries/2024-10-12-easy-to-hard-generalization-scalable-alignment-beyond-human-supervision/", "published_at": "2024-10-12T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073f6579f5", "title": "Training Language Models to Self-Correct via Reinforcement Learning", "url": "https://fanpu.io/summaries/2024-10-12-training-language-models-to-self-correct-via-reinforcement-learning/", "published_at": "2024-10-12T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073eaaa001", "title": "Unsupervised Dense Information Retrieval with Contrastive Learning", "url": "https://fanpu.io/summaries/2024-10-07-unsupervised-dense-information-retrieval-with-contrastive-learning/", "published_at": "2024-10-07T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073db15a5b", "title": "Generate rather than Retrieve: Large Language Models are Strong Context Generators", "url": "https://fanpu.io/summaries/2024-10-05-generate-rather-than-retrieve-large-language-models-are-strong-context-generators/", "published_at": "2024-10-05T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073ba30bc7", "title": "ASQA: Factoid Questions Meet Long-Form Answers", "url": "https://fanpu.io/summaries/2024-10-03-asqa-factoid-questions-meet-long-form-answers/", "published_at": "2024-10-03T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073bf85585", "title": "Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergy", "url": "https://fanpu.io/summaries/2024-10-03-enhancing-retrieval-augmented-large-language-models-with-iterative-retrieval-generation-synergy/", "published_at": "2024-10-03T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073cc74619", "title": "Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions", "url": "https://fanpu.io/summaries/2024-10-03-interleaving-retrieval-with-chain-of-thought-reasoning-for-knowledge-intensive-multi-step-questions/", "published_at": "2024-10-03T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073b92bb34", "title": "Open-source Large Language Models are Strong Zero-shot Query Likelihood Models for Document Ranking", "url": "https://fanpu.io/summaries/2024-10-02-open-source-large-language-models-are-strong-zero-shot-query-likelihood-models-for-document-ranking/", "published_at": "2024-10-02T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073af9fe77", "title": "InPars: Data Augmentation for Information Retrieval using Large Language Models", "url": "https://fanpu.io/summaries/2024-09-29-inpars-data-augmentation-for-information-retrieval-using-large-language-models/", "published_at": "2024-09-29T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073b154c88", "title": "Lost in the Middle: How Language Models Use Long Contexts", "url": "https://fanpu.io/summaries/2024-09-29-lost-in-the-middle-how-language-models-use-long-contexts/", "published_at": "2024-09-29T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073af8b164", "title": "Precise Zero-Shot Dense Retrieval without Relevance Labels (HyDE)", "url": "https://fanpu.io/summaries/2024-09-27-precise-zero-shot-dense-retrieval-without-relevance-labels/", "published_at": "2024-09-27T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073ae00a3f", "title": "Dense X Retrieval: What Retrieval Granularity Should We Use?", "url": "https://fanpu.io/summaries/2024-09-23-dense-x-retrieval-what-retrieval-granularity-should-we-use/", "published_at": "2024-09-23T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073a4f23e2", "title": "Lift Yourself Up: Retrieval-augmented Text Generation with Self Memory", "url": "https://fanpu.io/summaries/2024-09-22-lift-yourself-up-retrieval-augmented-text-generation-with-self-memory/", "published_at": "2024-09-22T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073ac1927b", "title": "Query Rewriting for Retrieval-Augmented Large Language Models", "url": "https://fanpu.io/summaries/2024-09-22-query-rewriting-for-retrieval-augmented-large-language-models/", "published_at": "2024-09-22T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fffdd2c3", "title": "Mt Rainier | Fan Pu Zeng", "url": "https://fanpu.io/news/mt-rainier/", "published_at": "2024-08-26T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06ff4dd1b8", "title": "50th Paper Summary | Fan Pu Zeng", "url": "https://fanpu.io/news/50th-paper-summary/", "published_at": "2024-08-05T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0738ec321a", "title": "Deep Double Descent: Where Bigger Models and More Data Hurt", "url": "https://fanpu.io/summaries/2024-08-05-deep-double-descent-where-bigger-models-and-more-data-hurt/", "published_at": "2024-08-05T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07398339e0", "title": "Reconciling modern machine learning practice and the bias-variance trade-off", "url": "https://fanpu.io/summaries/2024-08-05-reconciling-modern-machine-learning-practice-and-the-bias-variance-trade-off/", "published_at": "2024-08-05T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07382964f1", "title": "Domain-Adjusted Regression or: ERM May Already Learn Features Sufficient for Out-of-Distribution Generalization", "url": "https://fanpu.io/summaries/2024-07-28-domain-adjusted-regression-or-erm-may-already-learn-features-sufficient-for-out-of-distribution-generalization/", "published_at": "2024-07-28T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073732bd61", "title": "Editing a classifier by rewriting its prediction rules", "url": "https://fanpu.io/summaries/2024-07-27-editing-a-classifier-by-rewriting-its-prediction-rules/", "published_at": "2024-07-27T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0736fad2aa", "title": "Distilling the Knowledge in a Neural Network", "url": "https://fanpu.io/summaries/2024-07-25-distilling-the-knowledge-in-a-neural-network/", "published_at": "2024-07-25T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0736883439", "title": "Constitutional AI: Harmlessness from AI Feedback", "url": "https://fanpu.io/summaries/2024-07-23-constitutional-ai-harmlessness-from-ai-feedback/", "published_at": "2024-07-23T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07361081bb", "title": "Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision", "url": "https://fanpu.io/summaries/2024-07-21-weak-to-strong-generalization-eliciting-strong-capabilities-with-weak-supervision/", "published_at": "2024-07-21T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07354cf0eb", "title": "Optimizing Instructions and Demonstrations for Multi-Stage Language Model Programs", "url": "https://fanpu.io/summaries/2024-07-14-optimizing-instructions-and-demonstrations-for-multi-stage-language-model-programs/", "published_at": "2024-07-14T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07345c6c9d", "title": "Large Language Models as Optimizers", "url": "https://fanpu.io/summaries/2024-07-13-large-language-models-as-optimizers/", "published_at": "2024-07-13T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0733994594", "title": "Prefix-Tuning: Optimizing Continuous Prompts for Generation", "url": "https://fanpu.io/summaries/2024-04-26-prefix-tuning-optimizing-continuous-prompts-for-generation/", "published_at": "2024-04-26T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073380f296", "title": "LoRA: Low-Rank Adaptation of Large Language Models", "url": "https://fanpu.io/summaries/2024-03-31-lora-low-rank-adaptation-of-large-language-models/", "published_at": "2024-03-31T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0732c5789d", "title": "LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens", "url": "https://fanpu.io/summaries/2024-03-27-longrope-extending-llm-context-window-beyond-2-million-tokens/", "published_at": "2024-03-27T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07327095e1", "title": "Scaling Laws for Neural Language Models", "url": "https://fanpu.io/summaries/2024-03-23-scaling-laws-for-neural-language-models/", "published_at": "2024-03-23T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0732afbc53", "title": "Training Compute-Optimal Large Language Models", "url": "https://fanpu.io/summaries/2024-03-23-training-compute-optimal-large-language-models/", "published_at": "2024-03-23T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07318b9713", "title": null, "url": "https://fanpu.io/summaries/2024-03-15-dspy-compiling-declarative-language-model-calls-into-self-improving-pipelines/", "published_at": "2024-03-15T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0731385d30", "title": null, "url": "https://fanpu.io/summaries/2024-02-22-colbert-efficient-and-effective-passage-search-via-contextualized-late-interaction-over-bert/", "published_at": "2024-02-22T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0730dfab29", "title": null, "url": "https://fanpu.io/summaries/2024-02-19-matryoshka-representation-learning/", "published_at": "2024-02-19T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0730888fb4", "title": null, "url": "https://fanpu.io/summaries/2023-12-12-large-language-models-for-software-engineering-survey-and-open-problems/", "published_at": "2023-12-12T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fed1851a", "title": null, "url": "https://fanpu.io/news/new-role/", "published_at": "2023-11-20T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073086651b", "title": null, "url": "https://fanpu.io/summaries/2023-11-03-high-resolution-image-synthesis-with-latent-diffusion-models/", "published_at": "2023-11-03T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a073001110b", "title": null, "url": "https://fanpu.io/summaries/2023-10-31-universal-and-transferable-adversarial-attacks-on-aligned-language-models/", "published_at": "2023-10-31T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fe9d8653", "title": null, "url": "https://fanpu.io/news/new-summary-pix2pix-zero/", "published_at": "2023-10-22T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072f76b03c", "title": null, "url": "https://fanpu.io/summaries/2023-10-22-zero-shot-image-to-image-translation/", "published_at": "2023-10-22T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072f1e1482", "title": null, "url": "https://fanpu.io/summaries/2023-10-17-instructpix2pix-learning-to-follow-image-editing-instructions/", "published_at": "2023-10-17T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072f15e407", "title": null, "url": "https://fanpu.io/summaries/2023-10-15-an-image-is-worth-one-word-personalizing-text-to-image-generation-using-textual-inversion/", "published_at": "2023-10-15T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072f153908", "title": null, "url": "https://fanpu.io/summaries/2023-10-04-on-the-dangers-of-stochastic-parrots-can-language-models-be-too-big/", "published_at": "2023-10-04T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072e5be9b3", "title": null, "url": "https://fanpu.io/summaries/2023-09-26-repository-level-prompt-generation-for-large-language-models-of-code/", "published_at": "2023-09-26T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072d5da91b", "title": null, "url": "https://fanpu.io/summaries/2023-09-24-rethinking-the-role-of-demonstrations-what-makes-in-context-learning-work/", "published_at": "2023-09-24T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072ccec15f", "title": null, "url": "https://fanpu.io/summaries/2023-09-23-calibrate-before-use-improving-few-shot-performance-of-language-models/", "published_at": "2023-09-23T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072cba8d2f", "title": null, "url": "https://fanpu.io/summaries/2023-09-10-understanding-deep-learning-requires-rethinking-generalization/", "published_at": "2023-09-10T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fe16192a", "title": null, "url": "https://fanpu.io/news/new-summary-implicit-bias-gd/", "published_at": "2023-09-09T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072be8341a", "title": null, "url": "https://fanpu.io/summaries/2023-09-09-the-implicit-bias-of-gradient-descent-on-separable-data/", "published_at": "2023-09-09T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072bcecef7", "title": null, "url": "https://fanpu.io/summaries/2023-09-07-gradient-descent-provably-optimizes-over-parameterized-neural-networks/", "published_at": "2023-09-07T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072bb0797d", "title": null, "url": "https://fanpu.io/summaries/2023-09-04-loss-landscapes-and-optimization-in-over-parameterized-non-linear-systems-and-neural-networks/", "published_at": "2023-09-04T00:00:00+00:00" }, { "id": "01a08776-ea0a-718b-9349-da21d095b4ad", "title": "Playing Sound Voltex at Home: Setting Up Unnamed SDVX Clone with the Yuancon SDVX Controller", "url": "https://fanpu.io/blog/2023/setting-up-yuancon-controller-sound-voltex/", "published_at": "2023-09-02T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fd3255b1", "title": null, "url": "https://fanpu.io/news/new-post-sdvx/", "published_at": "2023-09-02T00:00:00+00:00" }, { "id": "01a08776-ea0a-718b-9349-da21d12d4d7a", "title": "Creating Trackback Requests for Static Sites", "url": "https://fanpu.io/blog/2023/creating-trackback-requests/", "published_at": "2023-09-01T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fc922d75", "title": null, "url": "https://fanpu.io/news/new-post-trackback/", "published_at": "2023-09-01T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072bab701a", "title": null, "url": "https://fanpu.io/summaries/2023-08-29-extracting-training-data-from-large-language-models/", "published_at": "2023-08-29T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072b224ee1", "title": null, "url": "https://fanpu.io/summaries/2023-08-27-a-watermark-for-large-language-models/", "published_at": "2023-08-27T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fc5709a5", "title": null, "url": "https://fanpu.io/news/structured-state-space-sequence-model-summary/", "published_at": "2023-08-26T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072a96589d", "title": null, "url": "https://fanpu.io/summaries/2023-08-25-efficiently-modeling-long-sequences-with-structured-state-spaces/", "published_at": "2023-08-25T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072a43e494", "title": null, "url": "https://fanpu.io/summaries/2023-08-22-accurate-detection-of-wake-word-start-and-end-using-a-cnn/", "published_at": "2023-08-22T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072a763949", "title": null, "url": "https://fanpu.io/summaries/2023-08-22-exploring-the-limits-of-transfer-learning-with-a-unified-text-to-text-transformer/", "published_at": "2023-08-22T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fb803ff2", "title": null, "url": "https://fanpu.io/news/new-job/", "published_at": "2023-08-21T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0729c29f3f", "title": null, "url": "https://fanpu.io/summaries/2023-08-19-transformers-in-speech-processing-a-survey/", "published_at": "2023-08-19T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fab48f12", "title": null, "url": "https://fanpu.io/news/img-format-migration/", "published_at": "2023-08-11T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a06fb488261", "title": null, "url": "https://fanpu.io/news/interactive-simulacra-summary/", "published_at": "2023-08-11T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07289c74a6", "title": null, "url": "https://fanpu.io/summaries/2023-08-11-generative-agents-interactive-simulacra-of-human-behavior/", "published_at": "2023-08-11T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0728e33da9", "title": null, "url": "https://fanpu.io/summaries/2023-08-11-improving-language-understanding-by-generative-pre-training/", "published_at": "2023-08-11T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07298718e8", "title": null, "url": "https://fanpu.io/summaries/2023-08-11-metagpt-meta-programming-for-multi-agent-collaborative-framework/", "published_at": "2023-08-11T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a072624d87d", "title": null, "url": "https://fanpu.io/summaries/2023-08-10-dense-passage-retrieval-for-open-domain-question-answering/", "published_at": "2023-08-10T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0726f701ee", "title": null, "url": "https://fanpu.io/summaries/2023-08-10-language-models-are-unsupervised-multitask-learners/", "published_at": "2023-08-10T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a0727ca51b3", "title": null, "url": "https://fanpu.io/summaries/2023-08-10-simple-synthetic-data-reduces-sycophancy-in-large-language-models/", "published_at": "2023-08-10T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07252a5f41", "title": null, "url": "https://fanpu.io/summaries/2023-08-09-bart-denoising-sequence-to-sequence-pre-training-for-natural-language-generation-translation-and-comprehension/", "published_at": "2023-08-09T00:00:00+00:00" }, { "id": "01a0cfc8-83ee-70e5-8c70-9a07244164a7", "title": null, "url": "https://fanpu.io/summaries/2023-08-06-evaluating-large-language-models-trained-on-code/", "published_at": "2023-08-06T00:00:00+00:00" } ] posts Claim your blog
Back to fanpu.io
Blog · corpus.blog/blogs/fanpu.io/posts

fanpu.io

fanpu.io

2026

Graphical Bayesian Networks for Predicting Asset Covariancesoriginal ↗
19 Sept 2026
Improving Domain Adaptation of Transformer Models For Generating Reddit Commentsoriginal ↗
19 Sept 2026
Pseudo-determinism for Graph Streaming Problemsoriginal ↗
19 Sept 2026
Analysis of Symmetry and Conventions in Off-Belief Learning in Hanabioriginal ↗
19 Sept 2026
Blackjack traineroriginal ↗
19 Sept 2026
Fermi Estimatesoriginal ↗
19 Sept 2026
https://fanpu.io/games/flyjack/original ↗
19 Sept 2026
Parallelism analyzeroriginal ↗
19 Sept 2026
Poker Traineroriginal ↗
19 Sept 2026
流觞 · liú shāngoriginal ↗
19 Sept 2026
Xiangqi traineroriginal ↗
19 Sept 2026
Research Sprint Day 4 - ???original ↗
18 Sept 2026
Research Sprint Day 3 - Seed Noise and Transformer Baselinesoriginal ↗
12 Sept 2026
Research Sprint Day 2 - Rooflinesoriginal ↗
11 Sept 2026
Day 1 - Verifying the flash-linear-attention Gated DeltaNet kernel on GB10 (DGX Spark)original ↗
10 Sept 2026
Concise Introduction to Gated DeltaNetsoriginal ↗
9 Sept 2026

2025

Diffusion-LM Improves Controllable Text Generationoriginal ↗
8 Jun 2025
Neural Networks from Maximizing Rate Reductionoriginal ↗
10 May 2025
Redunet Post | Fan Pu Zengoriginal ↗
10 May 2025
ReduNet: A White-box Deep Network from the Principle of Maximizing Rate Reductionoriginal ↗
10 May 2025
DiLoCo: Distributed Low-Communication Training of Language Modelsoriginal ↗
5 May 2025
Iclr 2025 | Fan Pu Zengoriginal ↗
20 Apr 2025
An Intuitive Introduction to Gaussian Processesoriginal ↗
21 Jan 2025
Gaussian Process | Fan Pu Zengoriginal ↗
21 Jan 2025
Bounding Mixing Times of Markov Chains via the Spectral Gaporiginal ↗
12 Jan 2025
Markov Chain Mixing Times | Fan Pu Zengoriginal ↗
12 Jan 2025

2024

Neurips 2024 | Fan Pu Zengoriginal ↗
29 Nov 2024
RAGAS: Automated Evaluation of Retrieval Augmented Generationoriginal ↗
4 Nov 2024
Easy-to-Hard Generalization: Scalable Alignment Beyond Human Supervisionoriginal ↗
12 Oct 2024
Training Language Models to Self-Correct via Reinforcement Learningoriginal ↗
12 Oct 2024
Unsupervised Dense Information Retrieval with Contrastive Learningoriginal ↗
7 Oct 2024
Generate rather than Retrieve: Large Language Models are Strong Context Generatorsoriginal ↗
5 Oct 2024
ASQA: Factoid Questions Meet Long-Form Answersoriginal ↗
3 Oct 2024
Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergyoriginal ↗
3 Oct 2024
Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questionsoriginal ↗
3 Oct 2024
Open-source Large Language Models are Strong Zero-shot Query Likelihood Models for Document Rankingoriginal ↗
2 Oct 2024
InPars: Data Augmentation for Information Retrieval using Large Language Modelsoriginal ↗
29 Sept 2024
Lost in the Middle: How Language Models Use Long Contextsoriginal ↗
29 Sept 2024
Precise Zero-Shot Dense Retrieval without Relevance Labels (HyDE)original ↗
27 Sept 2024
Dense X Retrieval: What Retrieval Granularity Should We Use?original ↗
23 Sept 2024
Lift Yourself Up: Retrieval-augmented Text Generation with Self Memoryoriginal ↗
22 Sept 2024
Query Rewriting for Retrieval-Augmented Large Language Modelsoriginal ↗
22 Sept 2024
Mt Rainier | Fan Pu Zengoriginal ↗
26 Aug 2024
50th Paper Summary | Fan Pu Zengoriginal ↗
5 Aug 2024
Deep Double Descent: Where Bigger Models and More Data Hurtoriginal ↗
5 Aug 2024
Reconciling modern machine learning practice and the bias-variance trade-offoriginal ↗
5 Aug 2024
Domain-Adjusted Regression or: ERM May Already Learn Features Sufficient for Out-of-Distribution Generalizationoriginal ↗
28 Jul 2024
Editing a classifier by rewriting its prediction rulesoriginal ↗
27 Jul 2024
Distilling the Knowledge in a Neural Networkoriginal ↗
25 Jul 2024
Constitutional AI: Harmlessness from AI Feedbackoriginal ↗
23 Jul 2024
Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervisionoriginal ↗
21 Jul 2024
Optimizing Instructions and Demonstrations for Multi-Stage Language Model Programsoriginal ↗
14 Jul 2024
Large Language Models as Optimizersoriginal ↗
13 Jul 2024
Prefix-Tuning: Optimizing Continuous Prompts for Generationoriginal ↗
26 Apr 2024
LoRA: Low-Rank Adaptation of Large Language Modelsoriginal ↗
31 Mar 2024
LongRoPE: Extending LLM Context Window Beyond 2 Million Tokensoriginal ↗
27 Mar 2024
Scaling Laws for Neural Language Modelsoriginal ↗
23 Mar 2024
Training Compute-Optimal Large Language Modelsoriginal ↗
23 Mar 2024
https://fanpu.io/summaries/2024-03-15-dspy-compiling-declarative-language-model-calls-into-self-improving-pipelines/original ↗
15 Mar 2024
https://fanpu.io/summaries/2024-02-22-colbert-efficient-and-effective-passage-search-via-contextualized-late-interaction-over-bert/original ↗
22 Feb 2024
https://fanpu.io/summaries/2024-02-19-matryoshka-representation-learning/original ↗
19 Feb 2024

2023

https://fanpu.io/summaries/2023-12-12-large-language-models-for-software-engineering-survey-and-open-problems/original ↗
12 Dec 2023
https://fanpu.io/news/new-role/original ↗
20 Nov 2023
https://fanpu.io/summaries/2023-11-03-high-resolution-image-synthesis-with-latent-diffusion-models/original ↗
3 Nov 2023
https://fanpu.io/summaries/2023-10-31-universal-and-transferable-adversarial-attacks-on-aligned-language-models/original ↗
31 Oct 2023
https://fanpu.io/news/new-summary-pix2pix-zero/original ↗
22 Oct 2023
https://fanpu.io/summaries/2023-10-22-zero-shot-image-to-image-translation/original ↗
22 Oct 2023
https://fanpu.io/summaries/2023-10-17-instructpix2pix-learning-to-follow-image-editing-instructions/original ↗
17 Oct 2023
https://fanpu.io/summaries/2023-10-15-an-image-is-worth-one-word-personalizing-text-to-image-generation-using-textual-inversion/original ↗
15 Oct 2023
https://fanpu.io/summaries/2023-10-04-on-the-dangers-of-stochastic-parrots-can-language-models-be-too-big/original ↗
4 Oct 2023
https://fanpu.io/summaries/2023-09-26-repository-level-prompt-generation-for-large-language-models-of-code/original ↗
26 Sept 2023
https://fanpu.io/summaries/2023-09-24-rethinking-the-role-of-demonstrations-what-makes-in-context-learning-work/original ↗
24 Sept 2023
https://fanpu.io/summaries/2023-09-23-calibrate-before-use-improving-few-shot-performance-of-language-models/original ↗
23 Sept 2023
https://fanpu.io/summaries/2023-09-10-understanding-deep-learning-requires-rethinking-generalization/original ↗
10 Sept 2023
https://fanpu.io/news/new-summary-implicit-bias-gd/original ↗
9 Sept 2023
https://fanpu.io/summaries/2023-09-09-the-implicit-bias-of-gradient-descent-on-separable-data/original ↗
9 Sept 2023
https://fanpu.io/summaries/2023-09-07-gradient-descent-provably-optimizes-over-parameterized-neural-networks/original ↗
7 Sept 2023
https://fanpu.io/summaries/2023-09-04-loss-landscapes-and-optimization-in-over-parameterized-non-linear-systems-and-neural-networks/original ↗
4 Sept 2023
Playing Sound Voltex at Home: Setting Up Unnamed SDVX Clone with the Yuancon SDVX Controlleroriginal ↗
2 Sept 2023
https://fanpu.io/news/new-post-sdvx/original ↗
2 Sept 2023
Creating Trackback Requests for Static Sitesoriginal ↗
1 Sept 2023
https://fanpu.io/news/new-post-trackback/original ↗
1 Sept 2023
https://fanpu.io/summaries/2023-08-29-extracting-training-data-from-large-language-models/original ↗
29 Aug 2023
https://fanpu.io/summaries/2023-08-27-a-watermark-for-large-language-models/original ↗
27 Aug 2023
https://fanpu.io/news/structured-state-space-sequence-model-summary/original ↗
26 Aug 2023
https://fanpu.io/summaries/2023-08-25-efficiently-modeling-long-sequences-with-structured-state-spaces/original ↗
25 Aug 2023
https://fanpu.io/summaries/2023-08-22-accurate-detection-of-wake-word-start-and-end-using-a-cnn/original ↗
22 Aug 2023
https://fanpu.io/summaries/2023-08-22-exploring-the-limits-of-transfer-learning-with-a-unified-text-to-text-transformer/original ↗
22 Aug 2023
https://fanpu.io/news/new-job/original ↗
21 Aug 2023
https://fanpu.io/summaries/2023-08-19-transformers-in-speech-processing-a-survey/original ↗
19 Aug 2023
https://fanpu.io/news/img-format-migration/original ↗
11 Aug 2023
https://fanpu.io/news/interactive-simulacra-summary/original ↗
11 Aug 2023
https://fanpu.io/summaries/2023-08-11-generative-agents-interactive-simulacra-of-human-behavior/original ↗
11 Aug 2023
https://fanpu.io/summaries/2023-08-11-improving-language-understanding-by-generative-pre-training/original ↗
11 Aug 2023
https://fanpu.io/summaries/2023-08-11-metagpt-meta-programming-for-multi-agent-collaborative-framework/original ↗
11 Aug 2023
https://fanpu.io/summaries/2023-08-10-dense-passage-retrieval-for-open-domain-question-answering/original ↗
10 Aug 2023
https://fanpu.io/summaries/2023-08-10-language-models-are-unsupervised-multitask-learners/original ↗
10 Aug 2023
https://fanpu.io/summaries/2023-08-10-simple-synthetic-data-reduces-sycophancy-in-large-language-models/original ↗
10 Aug 2023
https://fanpu.io/summaries/2023-08-09-bart-denoising-sequence-to-sequence-pre-training-for-natural-language-generation-translation-and-comprehension/original ↗
9 Aug 2023
https://fanpu.io/summaries/2023-08-06-evaluating-large-language-models-trained-on-code/original ↗
6 Aug 2023
Next