https://citeark.co https://citeark.co/zh https://citeark.co/search https://citeark.co/zh/search https://citeark.co/requests https://citeark.co/zh/requests https://citeark.co/docs/agent-api https://citeark.co/zh/docs/agent-api https://citeark.co/docs/authentication https://citeark.co/zh/docs/authentication https://citeark.co/docs/cap https://citeark.co/zh/docs/cap https://citeark.co/docs/concepts https://citeark.co/zh/docs/concepts https://citeark.co/docs/guides https://citeark.co/zh/docs/guides https://citeark.co/docs https://citeark.co/zh/docs https://citeark.co/docs/local-reproduction https://citeark.co/zh/docs/local-reproduction https://citeark.co/docs/mcp https://citeark.co/zh/docs/mcp https://citeark.co/docs/quickstart https://citeark.co/zh/docs/quickstart https://citeark.co/docs/rate-limits https://citeark.co/zh/docs/rate-limits https://citeark.co/docs/roadmap https://citeark.co/zh/docs/roadmap https://citeark.co/docs/api/arxiv https://citeark.co/zh/docs/api/arxiv https://citeark.co/docs/api/badges https://citeark.co/zh/docs/api/badges https://citeark.co/docs/api/forks https://citeark.co/zh/docs/api/forks https://citeark.co/docs/api/health https://citeark.co/zh/docs/api/health https://citeark.co/docs/api https://citeark.co/zh/docs/api https://citeark.co/docs/api/objects https://citeark.co/zh/docs/api/objects https://citeark.co/docs/api/platform https://citeark.co/zh/docs/api/platform https://citeark.co/docs/api/processing https://citeark.co/zh/docs/api/processing https://citeark.co/docs/api/repositories https://citeark.co/zh/docs/api/repositories https://citeark.co/docs/api/runs https://citeark.co/zh/docs/api/runs https://citeark.co/docs/api/social https://citeark.co/zh/docs/api/social https://citeark.co/about https://citeark.co/zh/about https://citeark.co/contact https://citeark.co/zh/contact https://citeark.co/privacy https://citeark.co/zh/privacy https://citeark.co/terms https://citeark.co/zh/terms https://citeark.co/claims https://citeark.co/zh/claims https://citeark.co/runs https://citeark.co/zh/runs https://citeark.co/daily https://citeark.co/zh/daily https://citeark.co/r/ganwumeng/random-features-for-large-scale-kernel-machines--5metdd 2026-09-29T15:57:45.042Z https://citeark.co/zh/r/ganwumeng/random-features-for-large-scale-kernel-machines--5metdd 2026-09-29T15:57:45.042Z https://citeark.co/r/citeark/just-in-time-memory-learning-to-curate-task-adaptive-memory-for-llm-agents--15g4ws 2026-09-29T07:05:37.779Z https://citeark.co/zh/r/citeark/just-in-time-memory-learning-to-curate-task-adaptive-memory-for-llm-agents--15g4ws 2026-09-29T07:05:37.779Z https://citeark.co/r/citeark/kite-kv-invariant-transformer-expansion-for-efficient-agentic-llm-scaling--8nfw3u 2026-09-29T06:36:22.449Z https://citeark.co/zh/r/citeark/kite-kv-invariant-transformer-expansion-for-efficient-agentic-llm-scaling--8nfw3u 2026-09-29T06:36:22.449Z https://citeark.co/r/citeark/informed-masking-structure-aware-perturbation-for-reinforcement-learning-in-diff--94axkp 2026-09-28T06:32:39.196Z https://citeark.co/zh/r/citeark/informed-masking-structure-aware-perturbation-for-reinforcement-learning-in-diff--94axkp 2026-09-28T06:32:39.196Z https://citeark.co/r/citeark/you-only-need-2-3-of-the-chosen-experts-an-empirical-study-of-dynamic-expert-pru--a34ppx 2026-09-26T07:12:22.426Z https://citeark.co/zh/r/citeark/you-only-need-2-3-of-the-chosen-experts-an-empirical-study-of-dynamic-expert-pru--a34ppx 2026-09-26T07:12:22.426Z https://citeark.co/r/citeark/how-strongly-should-task-state-influence-an-llm-agent--8zd84a 2026-09-26T06:45:20.588Z https://citeark.co/zh/r/citeark/how-strongly-should-task-state-influence-an-llm-agent--8zd84a 2026-09-26T06:45:20.588Z https://citeark.co/r/citeark/kv-cobra-kv-cache-compression-via-co-optimized-bit-rank-allocation--0beowq 2026-09-25T07:46:46.712Z https://citeark.co/zh/r/citeark/kv-cobra-kv-cache-compression-via-co-optimized-bit-rank-allocation--0beowq 2026-09-25T07:46:46.712Z https://citeark.co/r/citeark/when-and-how-should-an-agent-clarify-cigask-teaching-llms-to-clarify-via-counter--iisl0l 2026-09-25T07:10:40.551Z https://citeark.co/zh/r/citeark/when-and-how-should-an-agent-clarify-cigask-teaching-llms-to-clarify-via-counter--iisl0l 2026-09-25T07:10:40.551Z https://citeark.co/r/ganwumeng/auto-encoding-variational-bayes--jl84uq 2026-09-25T04:07:05.485Z https://citeark.co/zh/r/ganwumeng/auto-encoding-variational-bayes--jl84uq 2026-09-25T04:07:05.485Z https://citeark.co/r/citeark/abstention-and-noise-filtering-two-missing-primitives-of-softmax-attention--jien37 2026-09-24T06:37:53.838Z https://citeark.co/zh/r/citeark/abstention-and-noise-filtering-two-missing-primitives-of-softmax-attention--jien37 2026-09-24T06:37:53.838Z https://citeark.co/r/citeark/aspire-asynchronous-batched-self-speculative-decoding-for-long-context-llm-infer--f43ppe 2026-09-22T06:48:37.698Z https://citeark.co/zh/r/citeark/aspire-asynchronous-batched-self-speculative-decoding-for-long-context-llm-infer--f43ppe 2026-09-22T06:48:37.698Z https://citeark.co/r/citeark/diagnosing-the-fact-grounding-gap-in-multi-hop-question-answering--arkkn9 2026-09-21T06:42:33.544Z https://citeark.co/zh/r/citeark/diagnosing-the-fact-grounding-gap-in-multi-hop-question-answering--arkkn9 2026-09-21T06:42:33.544Z https://citeark.co/r/ganwumeng/deep-residual-learning-for-image-recognition--6jnjh0 2026-09-20T13:23:09.980Z https://citeark.co/zh/r/ganwumeng/deep-residual-learning-for-image-recognition--6jnjh0 2026-09-20T13:23:09.980Z https://citeark.co/r/citeark/paperdoctor-evidence-grounded-and-actionable-feedback-for-scientific-papers-in-p--xmpjzq 2026-09-20T07:26:52.404Z https://citeark.co/zh/r/citeark/paperdoctor-evidence-grounded-and-actionable-feedback-for-scientific-papers-in-p--xmpjzq 2026-09-20T07:26:52.404Z https://citeark.co/r/citeark/beyond-token-local-imitation-reward-compatible-temporal-credit-assignment-for-on--7jye4o 2026-09-20T06:45:50.788Z https://citeark.co/zh/r/citeark/beyond-token-local-imitation-reward-compatible-temporal-credit-assignment-for-on--7jye4o 2026-09-20T06:45:50.788Z https://citeark.co/r/citeark/nameless-tokenization-a-lossless-tokenizer-level-defense-against-control-token-f--p9pmav 2026-09-19T06:28:06.664Z https://citeark.co/zh/r/citeark/nameless-tokenization-a-lossless-tokenizer-level-defense-against-control-token-f--p9pmav 2026-09-19T06:28:06.664Z https://citeark.co/r/citeark/when-tool-calls-succeed-but-workflows-fail-anomalies-at-the-agent-tool-boundary--so6e7v 2026-09-18T06:49:04.591Z https://citeark.co/zh/r/citeark/when-tool-calls-succeed-but-workflows-fail-anomalies-at-the-agent-tool-boundary--so6e7v 2026-09-18T06:49:04.591Z https://citeark.co/r/ganwumeng/flashattention-2-faster-attention-with-better-parallelism-and-work-partitioning--lpussw 2026-09-17T19:04:21.560Z https://citeark.co/zh/r/ganwumeng/flashattention-2-faster-attention-with-better-parallelism-and-work-partitioning--lpussw 2026-09-17T19:04:21.560Z https://citeark.co/r/ganwumeng/egalitarian-gradient-descent-a-simple-approach-to-accelerated-grokking--jlsbn2 2026-09-15T16:21:38.048Z https://citeark.co/zh/r/ganwumeng/egalitarian-gradient-descent-a-simple-approach-to-accelerated-grokking--jlsbn2 2026-09-15T16:21:38.048Z https://citeark.co/r/citeark/procedural-graphs-self-evolving-execution-structures-for-llm-agents--cbtk6h 2026-09-12T06:30:58.021Z https://citeark.co/zh/r/citeark/procedural-graphs-self-evolving-execution-structures-for-llm-agents--cbtk6h 2026-09-12T06:30:58.021Z https://citeark.co/r/citeark/knowledge-distillation-during-mid-training-favors-reasoning-over-factual-recall--dt61dt 2026-09-08T06:58:29.446Z https://citeark.co/zh/r/citeark/knowledge-distillation-during-mid-training-favors-reasoning-over-factual-recall--dt61dt 2026-09-08T06:58:29.446Z https://citeark.co/r/citeark/defense-as-skill-evolving-runtime-guard-skill-for-skill-augmented-agents--rl2sck 2026-09-08T06:18:04.833Z https://citeark.co/zh/r/citeark/defense-as-skill-evolving-runtime-guard-skill-for-skill-augmented-agents--rl2sck 2026-09-08T06:18:04.833Z https://citeark.co/r/citeark/generalization-dynamics-of-lm-pre-training--awr6et 2026-10-02T18:54:32.301Z https://citeark.co/zh/r/citeark/generalization-dynamics-of-lm-pre-training--awr6et 2026-10-02T18:54:32.301Z https://citeark.co/r/citeark/just-in-time-agent-memory-with-runtime-agentic-research--0lc4uw 2026-10-03T08:07:37.768Z https://citeark.co/zh/r/citeark/just-in-time-agent-memory-with-runtime-agentic-research--0lc4uw 2026-10-03T08:07:37.768Z https://citeark.co/r/ganwumeng/fourier-features-let-networks-learn-high-frequency-functions-in-low-dimensional--dejrwa 2026-10-01T12:31:07.637Z https://citeark.co/zh/r/ganwumeng/fourier-features-let-networks-learn-high-frequency-functions-in-low-dimensional--dejrwa 2026-10-01T12:31:07.637Z https://citeark.co/r/ganwumeng/simplifying-graph-convolutional-networks--jl607e 2026-09-30T21:24:17.465Z https://citeark.co/zh/r/ganwumeng/simplifying-graph-convolutional-networks--jl607e 2026-09-30T21:24:17.465Z https://citeark.co/r/ganwumeng/transformers-are-ssms-generalized-models-and-efficient-algorithms-through-struct--rs9v6u 2026-09-14T05:38:30.346Z https://citeark.co/zh/r/ganwumeng/transformers-are-ssms-generalized-models-and-efficient-algorithms-through-struct--rs9v6u 2026-09-14T05:38:30.346Z https://citeark.co/r/citeark/online-self-weighted-fine-tuning--iqyzxx 2026-09-07T06:36:35.574Z https://citeark.co/zh/r/citeark/online-self-weighted-fine-tuning--iqyzxx 2026-09-07T06:36:35.574Z https://citeark.co/r/citeark/longpibench-a-long-context-benchmark-for-prompt-injection--6rf5kr 2026-09-02T06:56:47.304Z https://citeark.co/zh/r/citeark/longpibench-a-long-context-benchmark-for-prompt-injection--6rf5kr 2026-09-02T06:56:47.304Z https://citeark.co/r/citeark/geometry-of-divergence-tracking-hidden-state-trajectories-for-adaptive-multi-tur--z06w8y 2026-09-04T06:34:56.862Z https://citeark.co/zh/r/citeark/geometry-of-divergence-tracking-hidden-state-trajectories-for-adaptive-multi-tur--z06w8y 2026-09-04T06:34:56.862Z https://citeark.co/r/citeark/agenticrag-r1-agentic-reinforcement-learning-with-stack-memory-for-multi-step-re--drrsza 2026-09-03T06:45:12.979Z https://citeark.co/zh/r/citeark/agenticrag-r1-agentic-reinforcement-learning-with-stack-memory-for-multi-step-re--drrsza 2026-09-03T06:45:12.979Z https://citeark.co/r/citeark/logo-token-level-dynamic-local-global-attention--k7rv0t 2026-09-03T06:28:14.759Z https://citeark.co/zh/r/citeark/logo-token-level-dynamic-local-global-attention--k7rv0t 2026-09-03T06:28:14.759Z https://citeark.co/r/citeark/finding-where-the-buck-stops-an-automated-failure-attribution-based-reflection-f--j878vv 2026-09-02T06:35:27.228Z https://citeark.co/zh/r/citeark/finding-where-the-buck-stops-an-automated-failure-attribution-based-reflection-f--j878vv 2026-09-02T06:35:27.228Z https://citeark.co/ganwumeng 2026-08-23T17:40:03.446Z https://citeark.co/zh/ganwumeng 2026-08-23T17:40:03.446Z https://citeark.co/daily/2026-09-29 2026-09-29T00:00:00.000Z https://citeark.co/zh/daily/2026-09-29 2026-09-29T00:00:00.000Z https://citeark.co/daily/2026-09-26 2026-09-26T00:00:00.000Z https://citeark.co/zh/daily/2026-09-26 2026-09-26T00:00:00.000Z https://citeark.co/daily/2026-09-25 2026-09-25T00:00:00.000Z https://citeark.co/zh/daily/2026-09-25 2026-09-25T00:00:00.000Z https://citeark.co/daily/2026-09-20 2026-09-20T00:00:00.000Z https://citeark.co/zh/daily/2026-09-20 2026-09-20T00:00:00.000Z https://citeark.co/daily/2026-09-08 2026-09-08T00:00:00.000Z https://citeark.co/zh/daily/2026-09-08 2026-09-08T00:00:00.000Z https://citeark.co/daily/2026-09-03 2026-09-03T00:00:00.000Z https://citeark.co/zh/daily/2026-09-03 2026-09-03T00:00:00.000Z https://citeark.co/daily/2026-09-02 2026-09-02T00:00:00.000Z https://citeark.co/zh/daily/2026-09-02 2026-09-02T00:00:00.000Z https://citeark.co/daily/2026-09-01 2026-09-01T00:00:00.000Z https://citeark.co/zh/daily/2026-09-01 2026-09-01T00:00:00.000Z https://citeark.co/daily/2026-08-28 2026-08-28T00:00:00.000Z https://citeark.co/zh/daily/2026-08-28 2026-08-28T00:00:00.000Z