{"id":"9e73b6db-f417-4085-ba08-d8db324cc679","arxiv_id":"2608.04128","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A programmable four-level trapped-ion qudit performs an entanglement-free Grover search (up to 94.5% success) and a single-system CHSH-type contextuality test (S = 2.816 ± 0.082) within one platform.","lead":"A single trapped barium ion runs two linked experiments as a programmable four-level quantum processor: a search that identifies its target state up to 94.5% of the time, and a contextuality test that scores S = 2.816, above the classical limit of 2 and near the quantum maximum of about 2.828. A generalist should read this to see one particle's internal energy levels hosting a real quantum algorithm and a foundational test of quantum mechanics in the same device.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"CHSH contextuality claim lacks operational-equivalence checks: if the same logical observable is not identical across the four circuits, S>2 does not witness contextuality.","rationale":"The reader rated CONDITIONAL with MODERATE confidence, and I agree with that verdict. Among the reader's listed concerns, operational equivalence is the most load-bearing because it is a precondition for the contextuality interpretation rather than a clarity issue. The q3 readout description is actually consistent with the level assignment if |q0⟩ and |q3⟩ are the two 6S1/2 states, as the readout sentence implies, so I would not press that point as a correctness objection. Leakage into unaddressed Zeeman states is an important experimental check, but the paper at least budgets it and shows single-pulse Rabi data; it does not threaten the internal logic of the claim. Operational equivalence, by contrast, is asserted only implicitly, and the noncontextuality inequality has no force without it. This concern can be settled by measurement tomography of the four circuits. If equivalence holds, the contextuality claim is supported; if not, the CHSH number is real but mislabeled. The Grover fidelity claim is independent of this concern. Since the issue is an addressable experimental verification rather than a proven failure, I keep the reader's CONDITIONAL verdict; no verdict change is needed.","tokens_in":13736,"tokens_out":14582,"duration_ms":140240,"concrete_test":"Reconstruct the effective POVM of each of the four CHSH measurement circuits by preparing a tomographically complete set of probe states (or at least the eigenstates of X and Z) and recording the final fluorescence statistics for each circuit. Compute the fidelity between the reconstructed 'Z' measurement in circuits 1 and 3 and between the reconstructed 'X' measurement in circuits 2 and 4. If the paired fidelities are not within 1−ε of unity, with ε set by the stated measurement error, operational equivalence fails and the CHSH violation cannot be interpreted as contextuality. Equivalently, compare single-setting marginals on fixed probe states across contexts; any statistically significant difference is a direct violation of the equivalence assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's second headline—a state-dependent CHSH-type contextuality violation with S=2.816±0.082—is only a contextuality witness if the four measurement circuits realize the same effective observables in every context. Eq. (6) converts final populations into ⟨ZZ⟩, ⟨ZX⟩, ⟨XZ⟩, and ⟨XX⟩, and Eq. (7) combines them. But circuits 1–4 in Fig. 2 are different pulse sequences; nothing in the text shows that the 'Z' appearing in circuit 1 is operationally the same POVM as the 'Z' in circuit 3, or that the 'X' in circuit 2 is the same as the 'X' in circuit 4. The paper asserts that 'phase-controlled rotations define the required measurement bases' but does not provide the pulse decompositions, calibration data, or equivalence tests. If pulse-area errors, phase offsets, or parameter drift differ between contexts, the nominal observables are not equivalent; a noncontextual hidden-variable model with context-dependent measurements can then reproduce S>2, so the measured violation would not demonstrate quantum contextuality. This is not a minor interpretive gloss: the central claim of the paper includes contextuality as an experimentally demonstrated non-classical resource. The same weakness does not affect the Grover result, which is a direct fidelity measurement against an absolute ideal.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a programmable four-level optical qudit encoded in a single 138Ba+ ion, controlled through phase-programmable optical rotations on four selected transitions (R01, R02, R13, R23). Using this platform, the authors implement an entanglement-free Grover search over four states, reporting target-state success probabilities up to 94.5±2.0% (average 90.1%), and a CHSH-type state-dependent contextuality test with a maximum |S| = 2.816±0.082 near θ ≈ 3π/4, approaching the Tsirelson bound 2√2. The main text, Methods, and Supplementary Information include population data, uncertainty budgets, noise characterization, and QuTiP simulations of the calibrated dynamics. The paper's headline claims are that the same programmable single-ion platform can benchmark both algorithmic performance and non-classical contextuality.","tokens_in":13834,"tokens_out":8377,"duration_ms":77220,"significance":"If the contextuality claim is supported, the paper would be a valuable demonstration that a single programmable trapped-ion qudit can host both an algorithmic benchmark and a foundational non-classicality witness, with both results measured against absolute external benchmarks (classical bound 2, Tsirelson bound 2√2, and ideal Grover success probability 1). The internal consistency of the data is a clear strength: Table S2 traces a smooth sinusoidal S(θ) with the reported maximum, and Table S3 gives a probability matrix whose rows sum to unity within plausible errors. The supplementary is also unusually honest, explicitly stating that the long-time Allan deviation exponent and the low-frequency PSD slope are not statistically resolved. However, the contextuality claim currently lacks the operational-equivalence evidence required for a valid noncontextuality inequality, and this is a load-bearing gap for the paper's second headline result.","major_comments":[{"comment":"The contextuality witness is only valid if the four measurement circuits realize operationally equivalent observables in every context. The manuscript asserts that 'phase-controlled rotations define the required measurement bases' but does not give the pulse decompositions of the four circuits, nor does it provide any equivalence test showing that the effective POVM implementing 'Z' in Circuit 1 is identical to that in Circuit 3, and likewise for 'X' in Circuits 2 and 4. Without such data, a noncontextual model with context-dependent measurements can reproduce S > 2, so the observed violation would not demonstrate quantum contextuality. Please add explicit gate decompositions and equivalence checks (for example, tomography of the effective observables in each context), and include the associated uncertainty in the error budget; otherwise the contextuality claim should be rephrased as a measurement-dependent correlation.","section":"§II B, Eqs. (6)–(7), Fig. 2"},{"comment":"The interpretation of all measured populations as populations of the four computational states assumes that laser pulses do not leak population into unaddressed Zeeman levels of 5D5/2 or into the 5D3/2 manifold. The only leakage evidence shown is a single Rabi trace for R02 in Fig. S1b, whereas the Grover and CHSH sequences concatenate many pulses on four different transitions. The text lists 'state leakage' and 'off-resonant excitation' in the error discussion, but it does not quantify leakage for the full synthesized sequences. Please provide per-sequence leakage checks (for example, measuring population outside the computational basis after representative full circuits) or otherwise bound the leakage contribution independently for each headline measurement.","section":"§II A, Eqs. (1)–(2), Table S1, Fig. S1b"},{"comment":"The description of the |q3⟩ population measurement is internally inconsistent as written. The text states that shelving the |q0⟩ population and detecting the unshelved fluorescence 'provides the |q3⟩ population'; without first mapping |q3⟩ onto |q0⟩, the unshelved fluorescence would include |q1⟩, |q2⟩, and |q3⟩. Since every reported probability depends on this readout, please clarify the mapping sequence for each of the four states, or give the explicit reconstruction formula, so that the population assignments in Table S3 and Figs. 3 and 6 are reproducible.","section":"§IV C, readout protocol"}],"minor_comments":[{"comment":"The interaction graph is connected, not 'fully connected'; the adjacency matrix has zero entries for the q0–q3 and q1–q2 pairs. Please use 'connected' rather than 'fully connected'.","section":"§II A, after Eq. (2)"},{"comment":"Several rows of the Grover probability matrix sum to more than 1, with the largest excess about 4.4% (the target |q3⟩ row sums to 1.044). Please comment on whether this reflects correlated systematic errors, a normalization issue, or an artifact of rounding.","section":"Table S3"},{"comment":"The main text attributes long-term limitations to slow laser-frequency and magnetic-field fluctuations, but the supplementary explicitly states that the long-time Allan deviation exponent is statistically indistinguishable from flat and that the low-frequency PSD slope is not resolved. Please align the main-text claim with the supplementary's stated inconclusiveness.","section":"Discussion and §V C"},{"comment":"The reference list contains incomplete author entries ([9] 'P. H. et al.' and [13] 'X. S. et al.') and a duplicate ([8] and [32]). Please correct these.","section":"References"},{"comment":"No data or code availability statement is provided. For a quantitative experimental paper relying on QuTiP simulations with a calibrated Hamiltonian, please state where raw data and simulation scripts can be obtained.","section":"Data availability"},{"comment":"The performance comparison with earlier trapped-ion (60%) and superconducting (89–98%) implementations should be caveated: different encodings, qubit counts, and error-correction overheads make a direct fidelity comparison of limited quantitative value.","section":"§II C and Discussion"}],"recommendation":"major_revision","confidential_remarks":"The Grover result appears solid and internally consistent, and the supplementary's noise analysis is commendably careful. The main risk is the contextuality claim: without operational-equivalence data, the violation of the CHSH-type inequality is not a demonstrated contextuality witness. I recommend inviting a revision that either supplies the equivalence checks or softens the contextual claim. The paper is otherwise within the journal's scope and likely fixable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the Grover part is solid and the combined platform is genuinely new; the contextuality part is over-sold until the authors show that their four measurement circuits realize the same effective observables. The paper deserves a serious referee, but the referee should push for added evidence and better disclosure.\n\nWhat is actually new: one programmable four-level 138Ba+ ion runs both an entanglement-free Grover search and a CHSH-type noncontextuality inequality. The search success probabilities (84.8–94.5%) are measured against the absolute ideal of 1, the CHSH parameter is measured against the classical bound 2 and the Tsirelson bound, and nothing is fitted to produce those headline numbers. The uncertainty budget is unusually candid, including explicit statements that the long-time Allan deviation exponent and the PSD slope are statistically unresolved. That honesty earns real credit.\n\nThe numbers hold up internally. Table S2 traces a smooth sinusoidal S(theta) with |S|max near 2.82 at theta about 3π/4, matching the text. The Grover probability matrix rows sum to 1 within stated errors. So the experimental work seems competently done.\n\nThe main soft spot is exactly what the stress-test flagged: the contextuality claim rests on operational equivalence. Circuits 1–4 are different pulse sequences, and nothing in the paper shows that the 'Z' in circuit 1 is the same POVM as the 'Z' in circuit 3, or that the 'X' in circuit 2 is the same as in circuit 4. Without calibration data or equivalence tests, a noncontextual model with context-dependent measurements can reproduce S>2. This is not a minor gloss — contextuality is half the title. The Grover result is unaffected.\n\nOther issues, in decreasing severity: the sign of S differs between the text (positive max near 3π/4) and Table S2 (negative at 2.31 rad); the readout sequence for |q3> says shelving q0 and then measuring unshelved population gives |q3>, which as written is not obviously valid; no raw data or code are provided; and the bibliography is garbled (refs 9 and 13, duplicate ref 32 = ref 8), with the closest prior trapped-ion contextuality work missing. None of these is fatal, but together they make the paper feel rushed at the edges.\n\nWho gets value: experimentalists working on qudit control, trapped-ion platforms for quantum algorithms, and people studying whether single-system contextuality is a useful resource metric. The paper is worth engaging with, but only after the authors answer the operational-equivalence question. I would send it to peer review, expecting heavy revision on the contextuality framing and a request for calibration data.","headline":"Credible trapped-ion qudit experiment with a solid Grover result and a contextuality claim that is under-defended: the operational-equivalence requirement needs real evidence before S=2.816 can be called a contextuality witness.","tokens_in":14636,"tokens_out":1447,"would_cite":true,"duration_ms":15062,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["03.67.-a","03.65.Ta"],"model":"deepseek-v4-flash","headline":"A single $^{138}\\mathrm{Ba}^{+}$ ion, programmed as a four-level optical qudit, runs an entanglement-free Grover search with up to $94.5\\pm2.0\\%$ success and violates a CHSH-type noncontextuality inequality up to $S=2.816\\pm0.082$, near…","keywords":["quantum contextuality","qudit","Grover search","trapped ion","optical qudit","CHSH inequality","coherent control","barium ion"],"falsifier":"Run the full Grover and CHSH pulse sequences, stop after the final pulse, and detect the populations of the unused magnetic sublevels — the $m_J=\\pm3/2,\\pm5/2$ levels of $5D_{5/2}$ and the $5D_{3/2}$ manifold. If their combined population is not below the roughly $0.2\\%$ leakage allowed in the error budget, the four-level description of the processor is not closed and the reported fidelities would need reinterpretation.","tokens_in":13320,"feed_emoji":"⚛️","tokens_out":14328,"duration_ms":111067,"temperature":0.7,"pith_summary":"Quantum search is usually expected to need entanglement between many qubits; this paper argues that a single trapped ion, viewed as a four-level optical qudit, can run it through coherent interference alone, with the marked state found up to $94.5\\pm2.0\\%$ of the time. Using the same programmable ion, the paper reports a CHSH-type noncontextuality inequality violated up to $S=2.816\\pm0.082$, close to the quantum limit $2\\sqrt{2}$. Together these results claim that one control toolbox, based on four phase-programmable optical rotations, supplies both an algorithmic benchmark and a measured non-classical resource in a single physical system. The wider point of interest is that the two features can now be compared on one platform, separating the role of coherent interference from that of entanglement.","feed_headline":"Single ion hits 94.5% Grover and S=2.816 contextuality violation","feed_subtitle":"One programmable barium-ion qudit runs a search and a contextuality test without any entangling gate.","key_machinery":"The central object is the four-level optical qudit, encoded in two $6S_{1/2}$ and two $5D_{5/2}$ Zeeman sublevels of a single $^{138}\\mathrm{Ba}^{+}$ ion, driven on the narrow 1762 nm electric-quadrupole transition. The mechanism that carries the argument is the elementary rotation $R_{ij}(\\theta,\\phi)=\\exp\\!\\left[-i\\theta/2\\left(e^{-i\\phi}|i\\rangle\\langle j|+e^{i\\phi}|j\\rangle\\langle i|\\right)\\right]$, restricted to the four selected transitions. Concatenating these embedded $SU(2)$ pulses generates the full $SU(4)$ group; the same pulse compiler builds the equal-superposition Hadamard, the oracle phase flip, and the diffusion operator $U_s=2|s\\rangle\\langle s|-I$ for Grover search, and the basis-changing rotations that define the four CHSH measurement contexts. The effective two-qubit encoding is what lets a single particle emulate a two-qubit register, so no entangling gate is needed.","core_discovery":"Restated on the paper's own terms: a single $^{138}\\mathrm{Ba}^{+}$ ion, with a four-state computational basis drawn from the $6S_{1/2}$ ground and $5D_{5/2}$ metastable manifolds, is a programmable four-dimensional quantum processor. Four allowed optical transitions — $R_{01}$, $R_{02}$, $R_{13}$, $R_{23}$ — generate arbitrary $SU(4)$ operations when combined as phase-controlled $SU(2)$ rotations. Under the mapping $|q_0\\rangle=|00\\rangle$, $|q_1\\rangle=|01\\rangle$, $|q_2\\rangle=|10\\rangle$, $|q_3\\rangle=|11\\rangle$, the qudit behaves as a two-qubit register, so Grover's algorithm can be run with a phase oracle and a diffusion reflection, without any entangling gate. The measured success probabilities for the four marked states are $(94.2\\pm1.6)\\%$, $(94.5\\pm2.0)\\%$, $(84.8\\pm3.3)\\%$, and $(86.8\\pm3.2)\\%$; the same pulse framework yields a CHSH parameter $S=2.816\\pm0.082$ near $\\theta=3\\pi/4$, violating the noncontextual bound $S\\le2$.","pith_inferences":["If leakage into the unaddressed Zeeman and $5D_{3/2}$ levels is confirmed below the error budget by direct tomography, the same platform could compare different transition geometries to test whether larger contextuality violations track higher Grover fidelity, a connection the paper raises but does not settle.","The effective two-qubit encoding suggests a concrete extension: compile the same four-rotation sequences for a real two-qubit processor and compare single-qudit versus entangling-gate search at fixed total error, which would quantify the hardware overhead the paper claims to save.","The reduced amplitude of the second CHSH maximum near $\\theta=7\\pi/4$ is attributed to accumulated decoherence; a shorter or compensated pulse sequence at that angle would test whether residual off-resonant coupling also contributes.","Because the CHSH test is state-dependent, a natural follow-up is to prepare different input states and verify that the same effective observables are operationally equivalent across contexts; without that check, the contextuality interpretation rests on the paper's implicit measurement-equivalence assumption."],"forward_implications":["A single-qudit implementation can execute Grover's algorithm at higher fidelity here than earlier trapped-ion two-qubit demonstrations, which reported roughly 60% success, and at a level comparable to superconducting implementations.","The four transitions $\\{R_{01},R_{02},R_{13},R_{23}\\}$ are a sufficient primitive set for arbitrary $SU(4)$ control, so other qudit algorithms can be compiled from the same calibrated pulses.","Contextuality can be witnessed in one particle through a CHSH-type inequality, so tests of non-classicality do not require spatial separation or entangling measurements.","The same physical platform can report both an algorithm's success probability and a contextuality violation, making it possible to look for a quantitative link between the two.","Resource overhead is reduced: an effective two-qubit search runs inside one ion, avoiding the calibration and crosstalk costs of multi-ion entangling gates."],"supporting_citations":[{"why":"Supplies the theoretical claim that Grover search can run in a single multilevel quantum system without entanglement, which the experiment implements.","marker":"[20]"},{"why":"Extends the single-system search framework and gives the circuit-level comparison the authors use to justify their qudit design.","marker":"[21]"},{"why":"Earlier trapped-ion Grover implementation with roughly 60% fidelity, the baseline the qudit result is compared against.","marker":"[25]"},{"why":"Superconducting-qubit Grover implementation reporting over 89% fidelity (up to 98% with error correction), the other main comparison.","marker":"[26]"},{"why":"Describes the ion-trap device, calibration procedures, and control/readout methods this experiment builds on.","marker":"[23]"},{"why":"Characterizes the optical barium ion qubit and its detection fidelity, underpinning the shelving-based readout.","marker":"[27]"}],"fun_headline_variants":["Entanglement-free Grover hits 94.5% in a single ion","Single trapped ion: S=2.816 contextuality, 94.5% Grover","One ion does Grover without entanglement and breaks contextuality","Four-level qudit in one ion: 94.5% Grover, S=2.816 contextuality"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole result stands on the assumption that the four states chosen as the computational basis are the only ones the control light populates in any measurable way, so the measured populations are not secretly diluted by population sitting in other magnetic sublevels.","fun_headline_variants_meta":{"raw":{"variants":["Entanglement-free Grover hits 94.5% in a single ion","Single trapped ion: S=2.816 contextuality, 94.5% Grover","One ion does Grover without entanglement and breaks contextuality","Four-level qudit in one ion: 94.5% Grover, S=2.816 contextuality"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00161,"raw_usage":{"total_tokens":6463,"prompt_tokens":1052,"completion_tokens":5411,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":668,"completion_tokens_details":{"reasoning_tokens":5320}},"tokens_in":668,"tokens_out":5411,"duration_ms":34525,"temperature":1.0,"reasoning_tokens":5320,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:47:02.162787+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the full Grover and CHSH pulse sequences, stop after the final pulse, and detect the populations of the unused magnetic sublevels — the $m_J=\\pm3/2,\\pm5/2$ levels of $5D_{5/2}$ and the $5D_{3/2}$ manifold. If their combined population is not below the roughly $0.2\\%$ leakage allowed in the error budget, the four-level description of the processor is not closed and the reported fidelities would need reinterpretation.","supporting_citations":[{"cited_title":"Lloyd, Quantum search without entanglement, Physi- cal Review A61, 010301 (1999)","cited_arxiv_id":null,"evidence_quote":"Extends the single-system search framework and gives the circuit-level comparison the authors use to justify their qudit design."},{"cited_title":"Dutta, A","cited_arxiv_id":null,"evidence_quote":"Earlier trapped-ion Grover implementation with roughly 60% fidelity, the baseline the qudit result is compared against."},{"cited_title":"Brickman, P","cited_arxiv_id":null,"evidence_quote":"Superconducting-qubit Grover implementation reporting over 89% fidelity (up to 98% with error correction), the other main comparison."},{"cited_title":"Dutta and M","cited_arxiv_id":null,"evidence_quote":"Describes the ion-trap device, calibration procedures, and control/readout methods this experiment builds on."},{"cited_title":"Pokharel and D","cited_arxiv_id":null,"evidence_quote":"Characterizes the optical barium ion qubit and its detection fidelity, underpinning the shelving-based readout."}],"review_version":2}