{"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":"/paper/disentangling-and-integrating-relational-and","title":"Disentangling and Integrating Relational and Sensory Information in Transformer Architectures","arxiv_id":"2405.16727","date":"2024-05-26","proceeding":null,"authors":["Awni Altabaa","John Lafferty"],"abstract":"Relational reasoning is a central component of generally intelligent systems, enabling robust and data-efficient inductive generalization. Recent empirical evidence shows that many existing neural architectures, including Transformers, struggle with tasks requiring relational reasoning. In this work, we distinguish between two types of information: sensory information about the properties of individual objects, and relational information about the relationships between objects. While neural attention provides a powerful mechanism for controlling the flow of sensory information between objects, the Transformer lacks an explicit computational mechanism for routing and processing relational information. To address this limitation, we propose an architectural extension of the Transformer framework that we call the Dual Attention Transformer (DAT), featuring two distinct attention mechanisms: sensory attention for directing the flow of sensory information, and a novel relational attention mechanism for directing the flow of relational information. We empirically evaluate DAT on a diverse set of tasks ranging from synthetic relational benchmarks to complex real-world tasks such as language modeling and visual processing. Our results demonstrate that integrating explicit relational computational mechanisms into the Transformer architecture leads to significant performance gains in terms of data efficiency and parameter efficiency.","url_abs":"https://arxiv.org/abs/2405.16727v2","url_pdf":"https://arxiv.org/pdf/2405.16727v2.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":"disentangling-and-integrating-relational-and","repo_url":"https://github.com/awni00/dual-attention","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"disentangling-and-integrating-relational-and","repo_url":"https://github.com/awni00/abstract_transformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"information-retrieval","task_name":"Information Retrieval"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"relational-reasoning","task_name":"Relational Reasoning"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"set","method_name":"SET"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2405.16727","atlas_url":"https://app.syntology.ai/?focus=2405.16727","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.16727"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/awni00/abstract_transformer","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/awni00/dual-attention","reach":null}],"summary":{"ran":9,"ran_fixture":1,"ran_draft_wrong":3,"ran_violates":1,"unverified":7},"by_repo_kind":{"official":{"samples":6,"ran":2,"repositories":1},"listed":{"samples":15,"ran":12,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"437880805ab90b88","entry":"Attention","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"437880805ab90b88"}},{"code_sha256_prefix":"a001e1fb0893402a","entry":"FeedForwardBlock","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a001e1fb0893402a"}},{"code_sha256_prefix":"27aa5362b0778651","entry":"PositionRelativeSymbolRetriever","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"27aa5362b0778651"}},{"code_sha256_prefix":"07d2253837bc4d3f","entry":"PositionalSymbolRetriever","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"07d2253837bc4d3f"}},{"code_sha256_prefix":"01777e072296c0bb","entry":"RMSNorm","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"01777e072296c0bb"}},{"code_sha256_prefix":"f75b1ab91cff5f54","entry":"RelationalAttention","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f75b1ab91cff5f54"}},{"code_sha256_prefix":"e96ef6f7fa7fb78b","entry":"RelationalAttention","repo":"awni00/dual-attention","repo_kind":"official","path":"dual_attention/dual_attention.py","file_url":"https://github.com/awni00/dual-attention/blob/HEAD/dual_attention/dual_attention.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e96ef6f7fa7fb78b"}},{"code_sha256_prefix":"41fd4f0515e6826a","entry":"RelativePositionalEncoding","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"41fd4f0515e6826a"}},{"code_sha256_prefix":"32699e73b61b32e9","entry":"SymbolicAttention","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"32699e73b61b32e9"}},{"code_sha256_prefix":"6dc1f541b73c3081","entry":"apply_rotary_emb","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6dc1f541b73c3081"}},{"code_sha256_prefix":"dd15a61ea3d0e604","entry":"apply_rotary_emb","repo":"awni00/dual-attention","repo_kind":"official","path":"dual_attention/dual_attention.py","file_url":"https://github.com/awni00/dual-attention/blob/HEAD/dual_attention/dual_attention.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"dd15a61ea3d0e604"}},{"code_sha256_prefix":"a73cb1ba115d3af9","entry":"compute_causal_mask","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a73cb1ba115d3af9"}},{"code_sha256_prefix":"f108c0c79bea8101","entry":"create_norm","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f108c0c79bea8101"}},{"code_sha256_prefix":"074ed3e1882dc275","entry":"get_activation_function","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"074ed3e1882dc275"}},{"code_sha256_prefix":"ed6c6aa45ef23326","entry":"Attention","repo":"awni00/dual-attention","repo_kind":"official","path":"dual_attention/dual_attention.py","file_url":"https://github.com/awni00/dual-attention/blob/HEAD/dual_attention/dual_attention.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ed6c6aa45ef23326"}},{"code_sha256_prefix":"65050ac2515e807a","entry":"DisentangledRelationalCrossAttention","repo":"awni00/dual-attention","repo_kind":"official","path":"dual_attention/dual_attention.py","file_url":"https://github.com/awni00/dual-attention/blob/HEAD/dual_attention/dual_attention.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"65050ac2515e807a"}},{"code_sha256_prefix":"3cabca1c79c7e81e","entry":"DualAttention","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3cabca1c79c7e81e"}},{"code_sha256_prefix":"906b56a4d10f246c","entry":"DualAttention","repo":"awni00/dual-attention","repo_kind":"official","path":"dual_attention/dual_attention.py","file_url":"https://github.com/awni00/dual-attention/blob/HEAD/dual_attention/dual_attention.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"906b56a4d10f246c"}},{"code_sha256_prefix":"85bb295631648984","entry":"DualAttnEncoderBlock","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"85bb295631648984"}},{"code_sha256_prefix":"81995fb31f7a5e45","entry":"DualAttnTransformerLM","repo":"awni00/abstract_transformer","repo_kind":"listed","path":"dual_attention_transformer.py","file_url":"https://github.com/awni00/abstract_transformer/blob/HEAD/dual_attention_transformer.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"81995fb31f7a5e45"}},{"code_sha256_prefix":"8f6de9b1d93efd34","entry":"RelationalCrossAttention","repo":"awni00/dual-attention","repo_kind":"official","path":"dual_attention/dual_attention.py","file_url":"https://github.com/awni00/dual-attention/blob/HEAD/dual_attention/dual_attention.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8f6de9b1d93efd34"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}