{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/code/init-to-zero","entry":"init_to_zero","source":"Syntology graph, per-sample; not an archive number","read_at":"2026-09-24T18:15:14+00:00","claim":"Names are grouped by exact entry-name string. Same-named routines are NOT asserted to be equivalent; 'ran' means executed on a synthesized fixture, not correctness. n_samples_ran = sum of by_status over every status except 'unverified' (ran_draft_wrong and ran_fixture are failures of Syntology's instrument, not of the code); n_papers_ran = papers with at least one such sample.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"},"n_papers":17,"n_papers_ran":13,"units":"n_samples, n_samples_ran, n_samples_fingerprinted and by_status count distinct code bodies (code_sha256); n_places and n_places_pointer_only count places, one per (paper, code body) pair, which is also the unit of the samples list","n_samples":2,"n_samples_ran":1,"n_samples_fingerprinted":0,"n_places":17,"n_places_pointer_only":7,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":0,"ran_fixture":0,"ran":1,"unverified":1},"syntology":{"atlas_url":null,"mcp":null,"mcp_per_sample":{"tool":"get_code","arguments_in":"samples[].mcp_get_code"},"developers":"https://syntology.ai/developers"},"samples":[{"arxiv_id":"2605.15422","paper":"/paper/arxiv-2605-15422","title":"DualKV: Shared-Prompt Flash Attention for Efficient RL Training with Large Rollouts and Long Contexts","date":null,"month_inferred_from_arxiv_id":"2026-05","title_source":"syntology","repo":"amazon-science/dualkv-flash-attn-for-rl","path":"flash-attention/flash_attn/flash_attn_triton.py","file_url":"https://github.com/amazon-science/dualkv-flash-attn-for-rl/blob/HEAD/flash-attention/flash_attn/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2604.23798","paper":"/paper/arxiv-2604-23798","title":"ELSA: Exact Linear-Scan Attention for Fast and Memory-Light Vision Transformers","date":null,"month_inferred_from_arxiv_id":"2026-04","title_source":"syntology","repo":"Dao-AILab/flash-attention","path":"flash_attn/flash_attn_triton.py","file_url":"https://github.com/Dao-AILab/flash-attention/blob/HEAD/flash_attn/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2604.07658","paper":"/paper/arxiv-2604-07658","title":"Optimal Decay Spectra for Linear Recurrences","date":null,"month_inferred_from_arxiv_id":"2026-04","title_source":"syntology","repo":"SiLifen/PoST","path":"post_kernels/ssd_chunk_state.py","file_url":"https://github.com/SiLifen/PoST/blob/HEAD/post_kernels/ssd_chunk_state.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"f7a500c0f491fa7a","mcp_get_code":{"code_sha256":"f7a500c0f491fa7a"}},{"arxiv_id":"2601.06891","paper":"/paper/arxiv-2601-06891","title":"Contrastive Language-Image Mamba Pretraining","date":null,"month_inferred_from_arxiv_id":"2026-01","title_source":"syntology","repo":"NimrodShabtay/CLIMP","path":"models/vmamba/mamba2/ssd_bmm.py","file_url":"https://github.com/NimrodShabtay/CLIMP/blob/HEAD/models/vmamba/mamba2/ssd_bmm.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"f7a500c0f491fa7a","mcp_get_code":{"code_sha256":"f7a500c0f491fa7a"}},{"arxiv_id":"2505.00598","paper":"/paper/fast-and-low-cost-genomic-foundation-models","title":"Fast and Low-Cost Genomic Foundation Models via Outlier Removal","date":"2025-05-01","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"MAGICS-LAB/GERM","path":"DNABERT2/flash_attn_triton.py","file_url":"https://github.com/MAGICS-LAB/GERM/blob/HEAD/DNABERT2/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2503.14493","paper":"/paper/state-space-model-meets-transformer-a-new-1","title":"State Space Model Meets Transformer: A New Paradigm for 3D Object Detection","date":"2025-03-18","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"OpenSpaceAI/DEST3D","path":"issm_triton/issm_chunk_scan.py","file_url":"https://github.com/OpenSpaceAI/DEST3D/blob/HEAD/issm_triton/issm_chunk_scan.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"f7a500c0f491fa7a","mcp_get_code":{"code_sha256":"f7a500c0f491fa7a"}},{"arxiv_id":"2411.12537","paper":"/paper/unlocking-state-tracking-in-linear-rnns","title":"Unlocking State-Tracking in Linear RNNs Through Negative Eigenvalues","date":"2024-11-19","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"automl/unlocking_state_tracking","path":"mamba_dev/mamba_ssm/ops/triton/ssd_chunk_state.py","file_url":"https://github.com/automl/unlocking_state_tracking/blob/HEAD/mamba_dev/mamba_ssm/ops/triton/ssd_chunk_state.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"f7a500c0f491fa7a","mcp_get_code":{"code_sha256":"f7a500c0f491fa7a"}},{"arxiv_id":"2407.09941","paper":"/paper/hydra-bidirectional-state-space-models","title":"Hydra: Bidirectional State Space Models Through Generalized Matrix Mixers","date":"2024-07-13","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"goombalab/hydra","path":"hydra/bert/src/flash_attn_triton.py","file_url":"https://github.com/goombalab/hydra/blob/HEAD/hydra/bert/src/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2407.08608","paper":"/paper/flashattention-3-fast-and-accurate-attention","title":"FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision","date":"2024-07-11","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"dao-ailab/flash-attention","path":"flash_attn/flash_attn_triton.py","file_url":"https://github.com/dao-ailab/flash-attention/blob/HEAD/flash_attn/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2407.05483","paper":"/paper/just-read-twice-closing-the-recall-gap-for","title":"Just read twice: closing the recall gap for recurrent language models","date":"2024-07-07","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"HazyResearch/prefix-linear-attention","path":"based/ops/triton/linear.py","file_url":"https://github.com/HazyResearch/prefix-linear-attention/blob/HEAD/based/ops/triton/linear.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2402.18376","paper":"/paper/tokenization-is-more-than-compression","title":"Tokenization Is More Than Compression","date":"2024-02-28","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"kensho-technologies/timtc_vocabs_models","path":"models/1.3b/bpe_merge_3_40960/flash_attn_triton.py","file_url":"https://github.com/kensho-technologies/timtc_vocabs_models/blob/HEAD/models/1.3b/bpe_merge_3_40960/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2402.13561","paper":"/paper/cognitive-visual-language-mapper-advancing","title":"Cognitive Visual-Language Mapper: Advancing Multimodal Comprehension with Enhanced Visual Knowledge Alignment","date":"2024-02-21","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"hitsz-tmg/cognitive-visual-language-mapper","path":"LLaVA/llava/model/language_model/mpt/flash_attn_triton.py","file_url":"https://github.com/hitsz-tmg/cognitive-visual-language-mapper/blob/HEAD/LLaVA/llava/model/language_model/mpt/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2402.08777","paper":"/paper/dnabert-s-learning-species-aware-dna","title":"DNABERT-S: Pioneering Species Differentiation with Species-Aware DNA Embeddings","date":"2024-02-13","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"magics-lab/dnabert_s","path":"evaluate/model_codes/flash_attn_triton.py","file_url":"https://github.com/magics-lab/dnabert_s/blob/HEAD/evaluate/model_codes/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2312.15685","paper":"/paper/what-makes-good-data-for-alignment-a","title":"What Makes Good Data for Alignment? A Comprehensive Study of Automatic Data Selection in Instruction Tuning","date":"2023-12-25","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"hkust-nlp/deita","path":"src/deita/alignment/flash_attn/triton_flash_attention.py","file_url":"https://github.com/hkust-nlp/deita/blob/HEAD/src/deita/alignment/flash_attn/triton_flash_attention.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2310.05869","paper":"/paper/hyperattention-long-context-attention-in-near","title":"HyperAttention: Long-context Attention in Near-Linear Time","date":"2023-10-09","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"amirzandieh/HyperAttention","path":"src/flash_attn_triton.py","file_url":"https://github.com/amirzandieh/HyperAttention/blob/HEAD/src/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2309.06180","paper":"/paper/2309-06180","title":"Efficient Memory Management for Large Language Model Serving with PagedAttention","date":"2023-09-12","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"ROCm/flash-attention","path":"flash_attn/flash_attn_triton.py","file_url":"https://github.com/ROCm/flash-attention/blob/HEAD/flash_attn/flash_attn_triton.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}},{"arxiv_id":"2304.11277","paper":"/paper/pytorch-fsdp-experiences-on-scaling-fully","title":"PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel","date":"2023-04-21","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"pytorch-labs/applied-ai","path":"kernels/needs_perf_help/fp8_rowwise_tma_persistent.py","file_url":"https://github.com/pytorch-labs/applied-ai/blob/HEAD/kernels/needs_perf_help/fp8_rowwise_tma_persistent.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"code_sha256_prefix":"1586b639a5abd0b0","mcp_get_code":{"code_sha256":"1586b639a5abd0b0"}}]}