{"id":"462cdbb7-db23-4955-9cac-c6514bff627d","arxiv_id":"2607.06823","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Twisted copresheaves on double categories are equivalent to cloven loosely discrete opfibrations via an elements construction, with a collage comparison to double barrels.","lead":"The paper defines twisted double functors, which swap tight and loose morphisms between double categories, and uses them to model loose copresheaves. It proves these match a new fibrational notion called loosely discrete opfibrations, giving tools for loose Yoneda theory and systems composition.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified","rationale":"The reader's identification of the cleavage package as the weakest assumption is accurate: ordinary discrete opfibrations need no extra data, so the LARI-plus-algebra requirement is a genuine cost of the categorification. That package is, however, fully spelled out, used exactly where needed for the comparison cells and the round-trip, and does not introduce circularity or free parameters. The object-level proofs close; the deferred 2-categorical and collage parts are explicitly marked. Consequently the ACCEPT / HIGH-confidence verdict stands.","tokens_in":59424,"tokens_out":374,"duration_ms":4312,"concrete_test":"Independently re-derive the naturality of the loose-to-tight composition comparisons of Tw(P) (Construction 6.3 and Proposition 6.5) from the algebra associativity cell µ of Definition 3.11 alone, without invoking the square-filling property of Lemma 3.15; if the pentagon fails to commute, the pseudo-inverse construction has a hidden dependence.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central object-level equivalence (Propositions 6.9 and 6.14, Corollary 6.19) between twisted copresheaves and cloven loosely discrete opfibrations is carefully constructed in both directions. The LARI-plus-pseudo-algebra package of Definition 3.11 is an explicit categorification cost that the authors flag and use precisely to obtain the comparison cells and round-tripping; it is not a hidden gap. The collage equivalence is correctly left as Conjecture 5.1, and the full 2-categorical story is deferred as stated. No internal inconsistency or unstated assumption undermines the strongest claim.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper introduces twisted double functors (lax in the tight-to-loose direction and at least pseudo in the loose-to-tight direction) as a model of morphisms that exchange tight and loose structure, and uses them to define twisted copresheaves as normal lax functors into Prof. It constructs the twisted Hom, twisted representables, and many examples (self-indexing, subobjects, copresheaves, systems-theoretic actions). It then defines loosely discrete opfibrations as double functors for which the canonical map to the pullback of source along the object part is a weak equivalence of categories, equips them with cleavages that are LARI split equivalences whose target composite is a pseudo-algebra for the pull-push monad, and proves an object-level equivalence with twisted copresheaves via an elements construction and its pseudo-inverse (Propositions 6.9, 6.14; Corollary 6.19). A parallel comparison of twisted bimodules with double barrels via sections and collage is developed, with the full equivalence left as Conjecture 5.1.","tokens_in":59618,"tokens_out":891,"duration_ms":9929,"significance":"If the object-level correspondence holds as claimed, the paper supplies the missing foundational language for a loose Yoneda theory, for parameterized loose adjunctions needed in compact closure of double categories of relations/profunctors, and for the double-operadic organization of open systems already used in applied work. The constructions are explicit, the cleavage algebra is stated carefully rather than hidden, and the collage equivalence is correctly flagged as a conjecture. The work therefore opens a coherent research program rather than merely renaming existing notions.","major_comments":[{"comment":"The central claim is only an object-level equivalence (Propositions 6.9 and 6.14, Corollary 6.19). The 2-categorical structure of twisted functors (natural transformations, modifications, relative modifications) is developed in Section 4, yet the elements correspondence is never lifted to morphisms or 2-cells; Remark 4.7 explicitly notes that induced maps of opfibrations need not preserve cleavages. For a journal paper whose abstract advertises an equivalence of models, either a 2-functorial statement (even if only for a restricted class of morphisms) or a sharper delimitation of the claim is needed.","section":null},{"comment":"Conjecture 5.1 asserts that Sec and Col are mutually inverse 2-functors between double barrels and twisted bimodules. The constructions are given in detail (5.2–5.13), but the coherence argument that recovers the original comparison cells of a twisted bimodule after round-tripping is only sketched. Either a complete proof of the conjecture or a reduction of the remaining coherence to a known principle for twisted functors should be supplied; otherwise the comparison of models remains incomplete.","section":null}],"minor_comments":[{"comment":"Definition 2.1 is long; a short summary table of which comparison cells are required to be invertible for each variant (doubly lax / lax / normal / unitary / pseudo) would help the reader.","section":null},{"comment":"In Construction 3.18 the well-definition equation for cells is given only diagrammatically; the component form in Remark 3.19 is clearer and could be promoted to the main text.","section":null},{"comment":"Several applications (loose compact closure, systems theory) are motivational only; a single fully worked example that recovers a known adjunction or wiring-diagram action would strengthen the introduction.","section":null},{"comment":"Typographical consistency: “proarrow” vs “pro-arrow”, and occasional missing spaces after punctuation in long pasting diagrams.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The manuscript is solid foundational work and fits a pure category-theory journal. The main risk is that the deferred 2-categorical story and the open collage conjecture leave the paper feeling like the first half of a longer project; a clear statement of what is proved versus what is left open will help the editor place it."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This paper supplies the missing language for loose (proarrow-acting) copresheaves on double categories. The new pieces are twisted double functors (Def 2.1), loosely discrete opfibrations (Def 3.1), relative modifications (Def 4.16), and the object-level elements correspondence that equates twisted copresheaves with cloven loosely discrete opfibrations (Props 6.9, 6.14; Cor 6.19). That correspondence is the load-bearing result, and it is written carefully in both directions.\n\nWhat works well: the twisted Hom and the twisted representables are concrete and immediately usable (self-indexing of spans, subobjects, copresheaves into ADJ). Precomposition, the 2-category of twisted copresheaves, and the sections functor from double barrels are clean. The authors flag the LARI-plus-pseudo-algebra package on cleavages (Def 3.11) as the categorification cost that supplies comparison cells and round-tripping; they do not hide it. The collage equivalence is correctly left as Conjecture 5.1, and the full 2-categorical story is deferred as stated. Citations to Paré, Grandis, Shulman, Cruttwell–Lambert–Pronk–Szyld, and Libkind–Myers are accurate and non-circular.\n\nSoft spots are real but proportionate. The cleavage data is heavier than for ordinary discrete opfibrations; that is the price of the weak equivalence in Cat. Right-wobbliness of the reverse double functor (Rem 6.15) is acknowledged and left open. No formal verification, no code. None of this undermines the object-level claim.\n\nThis is for people already working with double categories, equipments, or double-operadic systems. It is infrastructure, not a paradigm shift, but the infrastructure is needed for loose Yoneda, loose compact closure, and the systems-theory examples the authors flag. I would send it to referees; the math is solid enough to deserve the time. I would cite the definitions and the elements correspondence when I next need loose copresheaves.","headline":"Solid foundational infrastructure for loose copresheaves: new twisted functors, loosely discrete opfibrations, and a carefully proved object-level equivalence, with the collage story correctly left as a conjecture.","tokens_in":60204,"tokens_out":551,"would_cite":true,"duration_ms":8902,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["18N10","18D30","18A40"],"pacs":[],"model":"grok-4.5","headline":"Twisted double functors and loosely discrete opfibrations give two equivalent models of loose copresheaves on double categories.","keywords":["double categories","twisted double functors","loosely discrete opfibrations","twisted copresheaves","elements construction","double barrels","collage","profunctors"],"falsifier":"Exhibit a concrete double category B and a twisted normal lax functor B ↪ Prof whose category of elements fails to be loosely discrete, or a cloven loosely discrete opfibration whose associated twisted functor is not isomorphic (via invertible natural transformation) to the original copresheaf.","tokens_in":60317,"feed_emoji":"↪","tokens_out":1018,"duration_ms":10762,"temperature":0.7,"pith_summary":"Double categories have two directions of morphisms—tight (function-like) and loose (relation- or module-like). Standard double-categorical Yoneda theory tracks only tight morphisms. This paper builds the missing loose counterpart. It defines twisted double functors, maps that swap the two directions and carry composition comparisons both ways, and uses them to form twisted copresheaves valued in profunctors. The same idea appears fibrationally as loosely discrete opfibrations: double functors that are discrete only in the loose direction, up to equivalence of categories rather than isomorphism. The paper proves that twisted copresheaves and cloven loosely discrete opfibrations correspond via an elements construction and its pseudo-inverse. A parallel collage construction relates twisted bimodules to double barrels (double categories over the walking loose arrow). The result supplies the foundational language needed for a loose Yoneda lemma, for loose compact closure, and for indexing open systems by double categories of interfaces.","feed_headline":"Loose copresheaves on double categories finally defined","feed_subtitle":"Twisted functors and loosely discrete opfibrations are equivalent, unlocking loose Yoneda and systems theory","key_machinery":"The elements construction that turns a twisted copresheaf into a cloven loosely discrete opfibration, together with its pseudo-inverse that recovers a unitary twisted functor from any such opfibration (and the collage that sends twisted bimodules to double barrels).","core_discovery":"Profunctor-valued twisted normal lax functors (twisted copresheaves) on a double category B are equivalent to cloven loosely discrete opfibrations over B. Every twisted copresheaf arises, up to invertible strict natural transformation, as the twisted functor associated to a cloven loosely discrete opfibration; every such opfibration is weakly equivalent over B to the category of elements of its associated twisted functor. The same data are also equivalent to pseudo-algebras for the pull-push monad of B.","pith_inferences":["The same twisting pattern should apply to other asymmetric 2-dimensional structures (e.g., virtual double categories or multicategories), yielding analogous “twisted” notions of copresheaf and fibration.","Once the 2-categorical correspondence is fully developed, the slice 2-category of cloven loosely discrete opfibrations over B will inherit a model structure or factorization system mirroring the ordinary discrete-opfibration case.","The collage conjecture, if true, would give a concrete presentation of the free double category generated by a twisted bimodule, useful for computation in systems theory."],"forward_implications":["A genuine loose Yoneda embedding and lemma for double categories can now be stated, with twisted representables playing the role of ordinary representables.","Loose compact closure and *-autonomy for double categories can be axiomatized using twisted Hom and twisted adjunctions rather than ordinary ones.","Open dynamical systems and structured cospans become algebras for twisted representables of double categories of interfaces and wiring patterns.","Twisted bimodules and double barrels become interchangeable models of loose bimodules, once the collage–sections equivalence is completed.","An elementary theory of flatness and geometric morphisms for double toposes can begin from the pull-push algebras that classify loosely discrete opfibrations."],"fun_headline_variants":["Twisted copresheaves match cloven loosely discrete opfibrations","Elements construction equates twisted functors to loose opfibrations","Loose copresheaves arise as twisted normal lax functors","Twisted double functors yield equivalent loosely discrete opfibrations","Pull-push monads recover twisted copresheaves on double categories"],"cache_read_input_tokens":49280,"weakest_assumption_plain":"A loosely discrete opfibration must be equipped with a cleavage that is a left-adjoint-right-inverse split equivalence whose target composite is a full pseudo-algebra for the base pull-push monad; without that algebraic package the two sides of the correspondence do not round-trip.","fun_headline_variants_meta":{"raw":{"variants":["Twisted copresheaves match cloven loosely discrete opfibrations","Elements construction equates twisted functors to loose opfibrations","Loose copresheaves arise as twisted normal lax functors","Twisted double functors yield equivalent loosely discrete opfibrations","Pull-push monads recover twisted copresheaves on double categories"]},"model":"grok-4.5","effort":"low","cost_usd":0.006948,"raw_usage":{"total_tokens":1740,"prompt_tokens":777,"num_sources_used":0,"completion_tokens":92,"cost_in_usd_ticks":69480000,"prompt_tokens_details":{"text_tokens":777,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":871,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":777,"tokens_out":92,"duration_ms":8258,"temperature":1.0,"reasoning_tokens":871,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-10T20:33:35.125179+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Exhibit a concrete double category B and a twisted normal lax functor B ↪ Prof whose category of elements fails to be loosely discrete, or a cloven loosely discrete opfibration whose associated twisted functor is not isomorphic (via invertible natural transformation) to the original copresheaf.","supporting_citations":[],"review_version":1}