{"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":"/census/reshape-for-broadcast","entry":"reshape_for_broadcast","source":"Syntology differential census (groundwork/55, run v2_2026-09-22), per sample; not an archive number","census_date":"2026-09-22","battery_sha256":["b986f7e04d794a0d88ad4c5f32cf63ec3590b5deff0192150737bc5f1c0b4677"],"runner_sha256":["5a452d0e7c0da5b80771d1be2afe3572e253568cf5c5d59a0d08ebd663d00808"],"bucket_key":"positional (rank, kind, dtype) of each array argument; the argument name is not part of the key because the harness draws the shared array from (rank, kind) and casts it to the dtype, whatever the name","bucket_fields":["rank","kind","dtype"],"claim":"Implementations sharing this entry name were each run on one shared input fixed by the positional (rank, kind, dtype) of their array arguments (the bucket). A cluster is the set whose recorded output digest (sha256 of the output rounded to 6 decimals) is identical. Identical values to six decimals on the shared input are agreement on those inputs, not a statement about the whole domain and not a substitution claim.","n_implementations_compared":6,"n_papers":10,"n_buckets":1,"n_distinct_outputs":1,"n_class_bearing":0,"not_compared":{"not_run_on_shared_input":{"n":5,"by_error":{"AssertionError":4,"RuntimeError":1}},"no_array_argument_ran_on_own_fixture_arguments_only":{"n":0,"n_papers":0,"recorded_shared_digest_equals_own_fixture_digest":0},"output_not_digested_non_numeric":{"n":0,"by_type":{}}},"withdrawn_excluded":0,"code_page":"/code/reshape-for-broadcast","buckets":[{"bucket":[[2,"float","float32"],[3,"float","float32"]],"n_implementations_compared":6,"n_papers":10,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"d7bd5ceec8a69dcd","size":6,"n_papers":10,"shape":[1,4,8],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":0.0,"members_with_recorded_values":6,"torch_reference_conventions_with_this_digest":[],"values":[1.8026286363601685,-1.5337762832641602,0.47605323791503906,-0.6908870339393616,-0.9028494954109192,-1.0304712057113647,-0.6627609133720398,0.5728892087936401,0.5188759565353394,0.9591456651687622,-0.3971995711326599,-1.5683481693267822,-0.28853508830070496,2.322108745574951,-0.5147839784622192,0.5505285859107971,0.06639543920755386,-1.6228755712509155,0.4321114122867584,-0.4061858654022217,-0.02259422466158867,-1.180782437324524,-0.5368682742118835,0.983674168586731,-0.7956128120422363,0.43327441811561584,0.9449777603149414,1.0175182819366455,-1.225785732269287,1.274217963218689,-1.3222218751907349,-0.5941091179847717],"values_recorded":32,"members":[{"code_sha256_prefix":"5a639d78ada17fee","path":"llama/llama/model.py","papers":["2310.19698","2312.00025","2507.03738","2604.11628"],"paper_pages":[{"arxiv_id":"2310.19698","page":"/paper/when-do-prompting-and-prefix-tuning-work-a"},{"arxiv_id":"2312.00025","page":"/paper/secure-transformer-inference"},{"arxiv_id":"2507.03738","page":"/paper/flow-anchored-consistency-models"},{"arxiv_id":"2604.11628","page":"/paper/arxiv-2604-11628"}],"arg_sig_recorded":[["freqs_cis",2,"float","float32"],["x",3,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"f8f09f78b474ee7c","path":"sam2/modeling/memory_attention.py","papers":["2510.24195","2501.07256"],"paper_pages":[{"arxiv_id":"2510.24195","page":"/paper/arxiv-2510-24195"},{"arxiv_id":"2501.07256","page":"/paper/edgetam-on-device-track-anything-model"}],"arg_sig_recorded":[["freqs_cis",2,"float","float32"],["x",3,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"0dc13509bbb03398","path":"md4/models/diffusion/md4.py","papers":["2406.04329"],"paper_pages":[{"arxiv_id":"2406.04329","page":"/paper/simplified-and-generalized-masked-diffusion"}],"arg_sig_recorded":[["freqs_cis",2,"float","float32"],["x",3,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"bdd3734871332b6b","path":"Cogformer/attention.py","papers":["2411.07176"],"paper_pages":[{"arxiv_id":"2411.07176","page":"/paper/more-expressive-attention-with-negative"}],"arg_sig_recorded":[["freqs_cis",2,"float","float32"],["x",3,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"c7b1e7c860ab5a79","path":"examples/llama/torchtitan/models/llama/model.py","papers":["2406.16793"],"paper_pages":[{"arxiv_id":"2406.16793","page":"/paper/adam-mini-use-fewer-learning-rates-to-gain"}],"arg_sig_recorded":[["freqs_cis",2,"float","float32"],["x",3,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"ea2f1af72961a8a5","path":"torchtitan/models/llama3/model.py","papers":["2410.06511"],"paper_pages":[{"arxiv_id":"2410.06511","page":"/paper/torchtitan-one-stop-pytorch-native-solution"}],"arg_sig_recorded":[["freqs_cis",2,"float","float32"],["x",3,"float","float32"]],"scalar_args":{},"class_bearing":false}]}]}]}