{"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/contrastive-loss","entry":"contrastive_loss","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":11,"n_papers":18,"n_buckets":3,"n_distinct_outputs":7,"n_class_bearing":0,"not_compared":{"not_run_on_shared_input":{"n":7,"by_error":{"RuntimeError":4,"AssertionError":1,"IndexError":1,"ValueError":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/contrastive-loss","buckets":[{"bucket":[[2,"float","float32"]],"n_implementations_compared":6,"n_papers":12,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"c4f8f09479c11d10","size":5,"n_papers":11,"shape":[],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":0.0,"members_with_recorded_values":5,"torch_reference_conventions_with_this_digest":[],"values":[1.4272676706314087],"values_recorded":1,"members":[{"code_sha256_prefix":"4675779aec92d6c0","path":"CLIP-ViP/src/modeling/CLIP_ViP.py","papers":["2209.06430","2510.14741","2410.07157","2608.20382"],"paper_pages":[{"arxiv_id":"2209.06430","page":"/paper/clip-vip-adapting-pre-trained-image-text"},{"arxiv_id":"2510.14741","page":"/paper/arxiv-2510-14741"},{"arxiv_id":"2410.07157","page":"/paper/instructg2i-synthesizing-images-from"},{"arxiv_id":"2608.20382","page":"/paper/arxiv-2608-20382"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"35caa662e2a8cf43","path":"M2PT/model/llava_archPT.py","papers":["2312.03781","2409.15657"],"paper_pages":[{"arxiv_id":"2312.03781","page":"/paper/lite-mind-towards-efficient-and-versatile"},{"arxiv_id":"2409.15657","page":"/paper/mmpt-multimodal-prompt-tuning-for-zero-shot"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"83ada93fd84968cc","path":"LLaVA/llava/model/language_model/llava_llama.py","papers":["2402.18695","2602.19146"],"paper_pages":[{"arxiv_id":"2402.18695","page":"/paper/grounding-language-models-for-visual-entity"},{"arxiv_id":"2602.19146","page":"/paper/arxiv-2602-19146"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"f2e5f28608115fa6","path":"MarT/models/modeling_clip.py","papers":["2210.00312","2601.01526"],"paper_pages":[{"arxiv_id":"2210.00312","page":"/paper/multimodal-analogical-reasoning-over"},{"arxiv_id":"2601.01526","page":"/paper/arxiv-2601-01526"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"54554dd514ac8105","path":"fgclip2/model/strcs/fgclip2.py","papers":["2505.05071"],"paper_pages":[{"arxiv_id":"2505.05071","page":"/paper/fg-clip-fine-grained-visual-and-textual"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{},"class_bearing":false}]},{"output_sha":"7891d656fa5f30a4","size":1,"n_papers":1,"shape":[],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[1.5417473316192627],"values_recorded":1,"members":[{"code_sha256_prefix":"d1d7fb60f123d370","path":"src/transformers/models/align/modeling_align.py","papers":["2502.14830"],"paper_pages":[{"arxiv_id":"2502.14830","page":"/paper/middle-layer-representation-alignment-for"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{},"class_bearing":false}]}]},{"bucket":[[2,"float","float32"],[2,"float","float32"]],"n_implementations_compared":4,"n_papers":4,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"2399fcd7f479a73c","size":1,"n_papers":1,"shape":[],"dtype":"float32","type":"Tensor","finite":false,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[NaN],"values_recorded":1,"members":[{"code_sha256_prefix":"e5fdec8e06236e4d","path":"model/model.py","papers":["2606.13007"],"paper_pages":[{"arxiv_id":"2606.13007","page":"/paper/arxiv-2606-13007"}],"arg_sig_recorded":[["z_t",2,"float","float32"],["z_f",2,"float","float32"]],"scalar_args":{"tau":"0.1"},"class_bearing":false}]},{"output_sha":"272f78036db6721f","size":1,"n_papers":1,"shape":[],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[0.25],"values_recorded":1,"members":[{"code_sha256_prefix":"a0ad7c051c1dd85d","path":"utils/pair_selection_util.py","papers":["1805.00833"],"paper_pages":[{"arxiv_id":"1805.00833","page":"/paper/learnable-pins-cross-modal-embeddings-for"}],"arg_sig_recorded":[["f_emb",2,"float","float32"],["v_emb",2,"float","float32"]],"scalar_args":{"margin":"1.0","tau":"0.5"},"class_bearing":false}]},{"output_sha":"570d68f10e57e52b","size":1,"n_papers":1,"shape":[],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[0.023483743891119957],"values_recorded":1,"members":[{"code_sha256_prefix":"ba00312b3d5b70dc","path":"methods/contrastive.py","papers":["2002.05709"],"paper_pages":[{"arxiv_id":"2002.05709","page":"/paper/a-simple-framework-for-contrastive-learning"}],"arg_sig_recorded":[["x0",2,"float","float32"],["x1",2,"float","float32"]],"scalar_args":{"tau":"0.07","norm":"True"},"class_bearing":false}]},{"output_sha":"7207fa2607ab5b36","size":1,"n_papers":1,"shape":[],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[0.7305240631103516],"values_recorded":1,"members":[{"code_sha256_prefix":"7567a11b33565b39","path":"dual_ae_trainer.py","papers":["2006.08386"],"paper_pages":[{"arxiv_id":"2006.08386","page":"/paper/coala-co-aligned-autoencoders-for-learning"}],"arg_sig_recorded":[["z_audio",2,"float","float32"],["z_tag",2,"float","float32"]],"scalar_args":{"t":"0.5"},"class_bearing":false}]}]},{"bucket":[[3,"float","float32"],[3,"float","float32"],[3,"float","float32"],[3,"float","float32"]],"n_implementations_compared":1,"n_papers":2,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"74999fd28ab18ccc","size":1,"n_papers":2,"shape":[],"dtype":"float32","type":"Tensor","finite":false,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[NaN],"values_recorded":1,"members":[{"code_sha256_prefix":"bd248e252b9ee711","path":"hete/model.py","papers":["2310.18884","2606.10284"],"paper_pages":[{"arxiv_id":"2310.18884","page":"/paper/simple-and-asymmetric-graph-contrastive-1"},{"arxiv_id":"2606.10284","page":"/paper/arxiv-2606-10284"}],"arg_sig_recorded":[["projected_emd",3,"float","float32"],["v_emd",3,"float","float32"],["sampled_embeddings_u",3,"float","float32"],["sampled_embeddings_neg_v",3,"float","float32"]],"scalar_args":{"tau":"0.1","lambda_loss":"1.0"},"class_bearing":false}]}]}]}