{"id":"58d1c7e3-5cbd-462f-82da-035305dba16a","arxiv_id":"2412.20963","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Two new GPT-level definitions of indistinguishable particles recover bosons and fermions for quantum pairs and expose theory-dependent particle-type structures.","lead":"This paper proposes two formal ways to define indistinguishable particles in any general probabilistic theory, one based on orbits of symmetric states and one based on categorical constructions. It shows the definitions reproduce bosons and fermions for pairs of quantum systems and identifies particle types in classical, Boxworld, and Spekkens toy models.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The diagrammatic particle-type classification relies on a 'unique most refined decomposition' theorem that is asserted but not proven in the paper; the central construction of Eq. (43) is therefore not yet established.","rationale":"The reader's weakest assumption was swap-invariance of states, but the more immediate obstruction is that the diagrammatic particle-type classification depends on a uniqueness theorem that the paper explicitly postpones to a future version. This is a genuine hole in the current manuscript, and it is flagged by the paper itself. The orbit-based framework has a related underdetermination: the choice between option I and option II changes the classical GPT classification, so the framework does not yet deliver a unique answer there either. These concerns reinforce, rather than overturn, the reader's CONDITIONAL verdict: the quantum pair recovery is essentially correct apart from a boson/fermion labelling slip in Sec. 3.6, and the categorical machinery is standard, but the general definition of particle types is not yet a complete, self-contained result.","tokens_in":14598,"tokens_out":12959,"duration_ms":141700,"concrete_test":"Complete the proof of the unique most-refined decomposition claimed in Sec. 3.6, or exhibit a finite process theory (e.g., the classical 2-bit theory in its Karoubi envelope) in which the symmetrisation idempotent has two inequivalent most-refined decompositions. If no proof can be given, or if a counterexample exists, the particle types from Eq. (43) are not canonical and the diagrammatic framework's central claim fails as stated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The diagrammatic framework's central construction is not yet proven. Sec. 3.6 defines particle types as the direct-summand systems (S^n_A)_i obtained by splitting the symmetrisation idempotent into a sum of orthogonal idempotents Part_i (Eqs. 42-43), and takes as canonical the 'unique most refined' such decomposition. That uniqueness is asserted twice, but the proof is explicitly deferred: 'This fact will be demonstrated in a forthcoming second version of this work,' and later 'In a second version of this work, we will see that this decomposition cannot be refined further.' The entire diagrammatic classification therefore rests on a theorem that is not included in the manuscript. If multiple inequivalent most-refined decompositions exist, or if the claimed coarse-graining property fails, then 'particle type' as defined by Eq. (43) is not well-defined. The orbit-based framework does not compensate for this gap: Sec. 2.2 leaves the choice between symmetric extremal and extremal symmetric states open, and Sec. 2.3.2 shows these choices give different particle-type counts for the classical GPT. The paper is a credible research programme, but the central claim that it supplies a general, theory-independent notion of particle types is not yet supported by a complete proof.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes two frameworks for defining types of indistinguishable particles in general probabilistic theories (GPTs). In the first, orbit-based framework, particle types are equivalence classes of symmetric pure states under the subgroup of transformations that preserve the symmetric subspace; in quantum theory this recovers bosons and fermions for pairs. In the second, diagrammatic framework, particle types are obtained by splitting the operational symmetrisation idempotent into orthogonal idempotents in the Karoubi envelope and biproduct completion, again recovering bosons and fermions for pairs of quantum systems. The paper works through several examples (quantum theory, classical GPT, Boxworld, and Spekkens' toy model) and discusses limitations, deferring two key proofs to a future version.","tokens_in":14891,"tokens_out":29771,"duration_ms":300761,"significance":"If the deferred results can be supplied, the paper would provide a genuinely theory-independent language for particle types, extending earlier swap-experiment approaches that are limited to a narrow class of GPTs. The orbit-based treatment is fully worked out for the examples and the recovery of bosons and fermions in quantum theory is convincingly demonstrated via Schur-Weyl duality. The paper is clearly written and the examples are instructive. However, the diagrammatic half of the paper currently rests on an unproven uniqueness assertion, and the claimed generality is not yet supported for classical GPTs.","major_comments":[{"comment":"The diagrammatic definition of particle types rests on the asserted existence and uniqueness of the \"most refined\" decomposition of the symmetrisation idempotent into orthogonal idempotents. This assertion is explicitly deferred to a second version (\"This fact will be demonstrated in a forthcoming second version of this work\" and \"In a second version of this work, we will see that this decomposition cannot be refined further\"). Because Eq. (43) and the interpretation of (S^n_A)_i as particle types depend on this uniqueness, the central construction of the diagrammatic framework is not yet established. Please provide the proof, or a precise statement of the conditions (e.g., centrality or invariance under the reversible transformations) under which the decomposition is unique, and verify that the quantum decomposition Sym^2_Q = B(H_s) ⊕ B(H_a) satisfies those conditions.","section":"3.6, Eqs. (42)–(43)"},{"comment":"The diagrammatic framework does not yet apply to classical GPTs: the text states that the definition \"only really makes sense if the system A ⊗ ... ⊗ A cannot itself be (nontrivially) written as a direct sum decomposition,\" that classical theory is the key counterexample, and that formalising this case is left to a second version. Since classical GPTs are the most basic examples and the abstract promises a general notion of indistinguishable particles in GPTs, this gap needs to be addressed before the framework can be said to deliver a theory-independent classification.","section":"3.6"},{"comment":"The orbit-based definition is parameterised by an unresolved choice between Option I (extremal symmetric states) and Option II (symmetric extremal states). The two options are not equivalent in general: for the classical GPT, Option I gives two particle types (diagonal and off-diagonal symmetric states) while Option II gives one. Since the paper does not select a preferred option or provide a criterion, the predicted set of particle types is not uniquely determined by the GPT alone. Please either justify a canonical choice or present the framework as a family of definitions parameterised by this choice.","section":"2.2 and 2.3.2"},{"comment":"The claim that the diagrammatic construction recovers bosons and fermions for all finite-dimensional quantum systems is only illustrated for two qubits in §3.6 (Sym^2_B(C^2) = B(C^3) ⊕ B(C^1)); the general-dimensional statement in §3.4 is asserted without proof. Once the uniqueness theorem from §3.6 is supplied, please give the general argument and state the sense in which the two summands are the bosonic and fermionic sectors.","section":"3.4 and 3.6"}],"minor_comments":[{"comment":"In the paragraph after Eq. (5), \"pure quantum states satisfying Eq. (5)\" is imprecise because Eq. (5) defines the action of the swap on basis vectors; the intended condition is Pψ = ±ψ, which follows from P^2 = 1.","section":"2.1"},{"comment":"In Option I, the states 1/2([j]⊗[k]+[k]⊗[j]) for j ≠ k are extremal in the symmetric state space but are mixtures in the full state space; calling them \"pure states\" could confuse readers, so adding a clarifying remark would help.","section":"2.3.2"},{"comment":"In Eq. (35), the rightmost diagram is labelled \"Sym^n_A\" but the surrounding text and the left side concern Sym^3_A; this appears to be a typo.","section":"3.4"},{"comment":"The notation (S^n_A)_i is used before it is defined; please introduce it explicitly.","section":"3.6"},{"comment":"There are several typos, e.g., \"indistinguishablility\" (§2.1), \"simpliticy\" (§3.4), and \"Symn_A\" versus \"Sym3_A\" in Eq. (35); a careful proofreading pass is needed.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is best seen as a research programme rather than a finished theory: the orbit-based part is largely complete, but the diagrammatic part depends on two explicitly deferred proofs. If the authors can supply the uniqueness proof for the symmetrisation-idempotent decomposition and handle the classical case, the paper would be a valuable contribution. The editorial decision should weigh whether the journal accepts manuscripts with such deferred load-bearing proofs."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know this paper does something actually new: it gives a theory-independent way to define indistinguishable particle types that works for arbitrary GPTs, not just those with a swap experiment. The orbit-based definition in Sec. 2.2 is simple and clear, and the quantum recovery of bosons and fermions is essentially right. The diagrammatic approach via Karoubi envelope and biproducts is a neat idea and well-motivated, even if it is not fully carried through. The examples—classical, Boxworld, Spekkens—are refreshingly concrete and honestly presented, and the paper is upfront about what it leaves open. I take the authors seriously as thinkers.\n\nThe soft spots are real but not fatal to the programme. The stress-test is right: the diagrammatic classification of particle types rests on the claimed uniqueness of the most refined decomposition of the symmetrisation idempotent, and the proof is deferred to a 'second version of this work.' That means Eq. (43) is not yet established, and 'particle type' in that framework is not yet a theorem but a conjecture. The orbit-based approach does not suffer from this, but it has its own unresolved ambiguity: the two choices for what counts as a symmetric pure state (extremal symmetric vs symmetric extremal) can give different particle types in classical theory, as the paper itself shows. That is not a mistake, but it means the definition is not yet uniquely pinned down. Also, the qubit example in Sec. 3.6 mislabels the sectors: the qutrit is the bosonic (symmetric) sector and the qubit is the fermionic (antisymmetric) sector, not the other way around. Minor in the grand scheme, but it should be fixed.\n\nThe Boxworld and Spekkens analyses are asserted more than proved, so I would want derivations before trusting those counts. Still, the central conceptual contribution—that particle types can be defined as orbits under symmetry-preserving transformations, and that this recovers quantum statistics—holds up for what it claims.\n\nWho is this for? People working on GPTs and on the foundations of indistinguishability. It is a good reading-group paper because it will spark discussion about what 'indistinguishability' should mean. I would cite it as the starting point for a new line of work, but I would pair it with the caveat that the diagrammatic part is not yet complete.\n\nRecommendation: yes, send it to peer review. It deserves refereeing, and the referees should push the authors to either supply the uniqueness proof or clearly mark it as a conjecture. The paper is not ready to be accepted as-is, but it is exactly the kind of nonstandard contribution a serious editor should engage with.","headline":"A genuinely new orbit-based framework for particle types in GPTs, but the diagrammatic half is a promising sketch whose central uniqueness theorem is explicitly deferred.","tokens_in":15452,"tokens_out":2449,"would_cite":true,"duration_ms":25811,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81P05","18M05","81P16"],"pacs":["03.65.Ta","03.65.-w"],"model":"deepseek-v4-flash","headline":"The paper defines indistinguishable particle types in any general probabilistic theory by symmetry orbits or by splitting the symmetrisation process, and recovers bosons and fermions for quantum pairs.","keywords":["general probabilistic theories","indistinguishable particles","particle types","exchange symmetry","operational symmetrisation","Karoubi envelope","biproducts","process theories"],"falsifier":"Exhibit a finite process theory in which the operational symmetrisation idempotent admits two different most-refined decompositions into orthogonal idempotents: the paper's claim that there is a unique canonical decomposition would then fail, and the diagrammatic particle-type sectors would no longer be well-defined.","tokens_in":14385,"feed_emoji":"⚛️","tokens_out":12118,"duration_ms":115297,"temperature":0.7,"pith_summary":"This paper tries to establish that indistinguishability of particles is a general feature that can be defined in any general probabilistic theory (GPT), not only in quantum theory. Its first definition types indistinguishable particles by the orbits of symmetric pure states under transformations that preserve the swap symmetry; its second, diagrammatic definition types them by the direct-sum sectors obtained when the operational symmetrisation process is split in a completed process theory. In both approaches a pair of identical quantum systems yields exactly two types, bosons and fermions, so the paper offers a theory-independent backdrop against which quantum indistinguishability can be compared with other theories. Applying the same definitions elsewhere gives one or two types in a classical GPT depending on a choice of pure-state convention, no new types in Boxworld, and three types in an epistemic toy theory.","feed_headline":"Particle types emerge from symmetry orbits in any probabilistic theory","feed_subtitle":"Orbit and diagrammatic definitions both recover bosons and fermions for quantum pairs.","key_machinery":"The orbit quotient $\\mathcal{O}_{T^s_S} = S^2_{ex,s}/T^s_S$ (Eq. 13) is the central object of the first framework: it collects symmetric pure states into orbits under swap-preserving transformations, and each orbit is declared to be one type of indistinguishable particle. The central object of the second framework is the operational symmetrisation idempotent $\\mathrm{Sym} = \\frac12(\\mathrm{id}+\\mathrm{swap})$ (Eq. 23), an equal mixture of doing nothing and swapping; in the Karoubi envelope, a categorical completion in which every idempotent is forced to split into a system, $\\mathrm{Sym}$ splits through a system $\\mathrm{Sym}^n_A$, and the unique most refined decomposition of $\\mathrm{Sym}$ into orthogonal idempotents turns that system into a biproduct (direct sum) of particle-type systems. The first object carries the state-space classification; the second carries the process-theoretic classification and also defines a non-disturbing 'particle type measurement' on symmetrised systems.","core_discovery":"On the paper's own terms, a pair of indistinguishable particles is a pair whose every state is invariant under relabelling, $\\omega = P\\omega$ for the swap map $P$. Particle types are then the equivalence classes of symmetric pure states under the subgroup $T^s_S$ of transformations that commute with $P$ on the symmetric subspace, namely the orbit set $\\mathcal{O}_{T^s_S} = S^2_{ex,s}/T^s_S$. In the diagrammatic formulation, the equal mixture of doing nothing and swapping, $\\mathrm{Sym} = \\frac12(\\mathrm{id} + P)$, is an idempotent process; splitting it via the Karoubi envelope and decomposing it into a sum of orthogonal idempotents produces a biproduct decomposition $\\mathrm{Sym}^n_A = \\bigoplus_i (S^n_A)_i$, and each summand is one particle type. For two quantum systems the decomposition recovers the antisymmetric (fermion) and symmetric (boson) sectors, e.g. $B(\\mathbb{C}^2)\\otimes B(\\mathbb{C}^2)$ symmetrises to $B(\\mathbb{C}^3)\\oplus B(\\mathbb{C}^1)$. The paper presents these as the two different routes to the same intended notion of particle type, applicable to arbitrary GPTs.","pith_inferences":["An open question the paper does not settle is whether the orbit classification and the diagrammatic sector decomposition always agree; finding a GPT where the two counts differ would show that 'particle type' is formalism-dependent, and one would need an extra principle to select the physical notion.","The dependence of the orbit definition on choosing symmetric-extremal states versus extremal symmetric states (the two options in Sec. 2.2) suggests that a fully predictive theory of particle types needs an additional operational criterion to fix the pure-state convention; the classical GPT example, where the two choices give two types versus one, is a concrete place to test such a criterion.","Because the framework deliberately sets aside trajectory-based indistinguishability, an extension combining exchange symmetry with trajectory labels might be needed for GPTs that admit definite particle paths; the present definitions would then describe only the exchange-symmetric sector of that richer notion.","The sector decomposition suggests a way to define fusion rules for particle types in GPTs by composing symmetrised systems and reducing the resulting idempotents, which could connect indistinguishable-particle statistics in non-quantum theories to anyonic behaviour."],"forward_implications":["If the orbit definition is correct, the particle types of a GPT are fixed by the geometry of its symmetric pure states and by which transformations preserve swap symmetry; in any transitive GPT whose pure states form a single orbit, the types are counted by the splitting of that single orbit under $T^s_S$.","The diagrammatic definition makes particle type a measurable, non-disturbing property of a symmetrised system, because the idempotent decomposition $\\mathrm{Sym} = \\sum_i \\mathrm{Part}_i$ is interpreted as the outcomes of a particle-type measurement.","For quantum systems of more than two particles, the diagrammatic approach predicts additional sectors beyond bosons and fermions, which the paper reads as paraparticles.","In Boxworld, imposing indistinguishability does not create new particle types: the product and entangled orbits already present are merely restricted to symmetric states.","In the toy model, the orbit definition predicts exactly three indistinguishable-particle types, distinguished by whether the epistemic state contains 0, 2, or 4 symmetric ontic states."],"supporting_citations":[{"why":"Defines the GPT framework and Boxworld, the non-transitive theory whose particle-type behaviour appears in Sec. 2.3.3.","marker":"[12]"},{"why":"Provides the standard presentation of GPT states, effects, and transformations on which the orbit construction is built.","marker":"[13]"},{"why":"Supplies the diagrammatic process-theory formalism in which the second classification is stated.","marker":"[19]"},{"why":"Establishes the relation between biproduct completion and the Karoubi envelope that the symmetrisation decomposition relies on.","marker":"[28]"},{"why":"Introduces the earlier swap-experiment approach to particle types that the new definitions aim to generalise.","marker":"[20]"},{"why":"Characterises the class of GPTs in which a swap experiment is well defined, motivating the need for a broader definition.","marker":"[21]"},{"why":"Indicates that the swap-experiment class is probably narrow, consisting of Euclidean Jordan algebra state spaces, which motivates the orbit and idempotent definitions.","marker":"[22]"},{"why":"Identifies all reversible transformations in Boxworld, which is needed to compute its orbits and show no new particle types appear.","marker":"[24]"},{"why":"Defines the epistemic toy model whose symmetric pure states split into three orbits under the allowed transformations.","marker":"[25]"},{"why":"Gives the convexified GPT presentation of the toy model used in the example.","marker":"[26]"}],"fun_headline_variants":["Symmetry orbits define particle types in any GPT","Bosons and fermions from symmetry orbits in GPTs","Particle types as orbits of symmetric states in GPTs","Orbit and diagram routes both yield bosons and fermions","Indistinguishability: from GPTs to bosons and fermions"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that physical indistinguishability is fully captured by exchange symmetry of states: a pair of particles counts as indistinguishable exactly when swapping their labels leaves the state invariant, so only swap-invariant states and swap-preserving transformations enter the classification; if a GPT could distinguish particles by tracking trajectories or by any other means, this definition would classify the wrong objects.","fun_headline_variants_meta":{"raw":{"variants":["Symmetry orbits define particle types in any GPT","Bosons and fermions from symmetry orbits in GPTs","Particle types as orbits of symmetric states in GPTs","Orbit and diagram routes both yield bosons and fermions","Indistinguishability: from GPTs to bosons and fermions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000207,"raw_usage":{"total_tokens":1381,"prompt_tokens":909,"completion_tokens":472,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":525,"completion_tokens_details":{"reasoning_tokens":389}},"tokens_in":525,"tokens_out":472,"duration_ms":4696,"temperature":1.0,"reasoning_tokens":389,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T23:06:10.351718+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Exhibit a finite process theory in which the operational symmetrisation idempotent admits two different most-refined decompositions into orthogonal idempotents: the paper's claim that there is a unique canonical decomposition would then fail, and the diagrammatic particle-type sectors would no longer be well-defined.","supporting_citations":[{"cited_title":"Information processing in generalized probabilistic theories","cited_arxiv_id":null,"evidence_quote":"Defines the GPT framework and Boxworld, the non-transitive theory whose particle-type behaviour appears in Sec. 2.3.3."},{"cited_title":"Selby, Carlo Maria Scandolo, and Bob Coecke","cited_arxiv_id":null,"evidence_quote":"Supplies the diagrammatic process-theory formalism in which the second classification is stated."},{"cited_title":"Two roads to classicality","cited_arxiv_id":null,"evidence_quote":"Establishes the relation between biproduct completion and the Karoubi envelope that the symmetrisation decomposition relies on."},{"cited_title":"Generalised phase kick-back: the structure of computational algorithms from physical principles","cited_arxiv_id":null,"evidence_quote":"Characterises the class of GPTs in which a swap experiment is well defined, motivating the need for a broader definition."},{"cited_title":"All reversible dynamics in maximally nonlocal theories are trivial.Phys- ical review letters , 104(8):080402, 2010","cited_arxiv_id":null,"evidence_quote":"Identifies all reversible transformations in Boxworld, which is needed to compute its orbits and show no new particle types appear."},{"cited_title":"Spekkens","cited_arxiv_id":null,"evidence_quote":"Defines the epistemic toy model whose symmetric pure states split into three orbits under the allowed transformations."},{"cited_title":"Generalized probabilistic theories without the no-restriction hypothesis","cited_arxiv_id":null,"evidence_quote":"Gives the convexified GPT presentation of the toy model used in the example."}],"review_version":1}