{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/visualizing-and-understanding-neural-models","title":"Visualizing and Understanding Neural Models in NLP","arxiv_id":"1506.01066","date":"2015-06-02","proceeding":"NAACL 2016 6","authors":["Jiwei Li","Xinlei Chen","Eduard Hovy","Dan Jurafsky"],"abstract":"While neural networks have been successfully applied to many NLP tasks the\nresulting vector-based models are very difficult to interpret. For example it's\nnot clear how they achieve {\\em compositionality}, building sentence meaning\nfrom the meanings of words and phrases. In this paper we describe four\nstrategies for visualizing compositionality in neural models for NLP, inspired\nby similar work in computer vision. We first plot unit values to visualize\ncompositionality of negation, intensification, and concessive clauses, allow us\nto see well-known markedness asymmetries in negation. We then introduce three\nsimple and straightforward methods for visualizing a unit's {\\em salience}, the\namount it contributes to the final composed meaning: (1) gradient\nback-propagation, (2) the variance of a token from the average word node, (3)\nLSTM-style gates that measure information flow. We test our methods on\nsentiment using simple recurrent nets and LSTMs. Our general-purpose methods\nmay have wide applications for understanding compositionality and other\nsemantic properties of deep networks , and also shed light on why LSTMs\noutperform simple recurrent nets,","url_abs":"http://arxiv.org/abs/1506.01066v2","url_pdf":"http://arxiv.org/pdf/1506.01066v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"visualizing-and-understanding-neural-models","repo_url":"https://github.com/phnk/D7047E","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"negation","task_name":"Negation"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1506.01066","atlas_url":"https://app.syntology.ai/?focus=1506.01066","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"1506.01066"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"1506.01066\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"1506.01066"},"call":"get_concepts_for_paper(arxiv_id=\"1506.01066\")"}],"by_repo":[{"repo":"phnk/D7047E","kind":"listed","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"31c09ddac555f47e09363033e711097fd982b4323f88514599f001ceb1c50b20","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}