{"id":"cfea8535-5702-41de-b094-11004a1799f7","arxiv_id":"2501.01731","paper_version":3,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Tensor light shifts make it possible to selectively rotate any chosen pair of the ten nuclear spin states of 87Sr, enabling Ramsey interferometry, parallel multi-field sensing, and simultaneous measurement of non-commuting collective observables.","lead":"Physicists manipulated the nuclear spin of ultracold strontium-87 atoms, performing precise rotations between selected spin states and running interferometers that stay coherent for seconds. The work gives alkaline-earth atoms a practical toolbox for qudit-based quantum computing and for simulating quantum magnets with SU(N) symmetry.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Full-manifold su(10) control is extrapolated: only four of ten states are demonstrated, and the proposed b>|q|(2F-1) fix for the observed uncontrolled transfers is not tested.","rationale":"The reader's weakest assumption identifies exactly the load-bearing gap: the complete-control claim is supported on only four of ten Zeeman states, and the proposed magnetic-field fix is not demonstrated. The manuscript itself discloses this limitation in Sec. IV, which strengthens rather than weakens the concern: the authors state that uncontrolled transfers occurred when driving other states and that a higher field 'could lift this issue,' but they do not show it. The condition b > |q|(2F-1) is physically plausible, but plausibility is not demonstration, and the abstract/conclusion language goes beyond the data by claiming a fully exploitable su(10) resource. The other concerns noted by the reader (heuristic decoherence model, calibration of eta(-3/2), asserted mathematical claims) are secondary: they affect error budgets or derivations but do not threaten the core demonstrated coherent rotations and interferometry. The four-state results, including 3-s Ramsey coherence and simultaneous non-commuting observable readout, appear internally consistent and are a real experimental advance. I therefore keep the CONDITIONAL verdict, with the condition being the experimental validation of full-manifold control in the proposed high-field regime.","tokens_in":22544,"tokens_out":9031,"duration_ms":93075,"concrete_test":"Repeat the Rabi and Ramsey sequences at a bias field satisfying b > |q|(2F-1) = 8|q|, e.g., b/h ~ 3.2 kHz with q/h ~ -320 Hz (roughly B ~ 14 G instead of 5.2 G), and demonstrate Rabi oscillations with no measurable population leakage on all nine adjacent pairs, especially pairs crossing the level-crossing region such as -1/2 <-> +1/2 and +1/2 <-> +3/2, while detecting all ten spin populations. If fidelities remain above 0.99 and uncontrolled transfers disappear, the full-manifold su(10) claim is supported; if transfers persist or fidelities degrade, the full-toolkit claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline claim of a 'toolkit for fully exploiting' the 10-state manifold, and its Sec. VI assertion that the operations implement generators of su(10) with complete control, require that every adjacent pair of Zeeman states can be spectrally isolated by the tensor light shift. The data demonstrate this only for the three adjacent pairs within m_F in {-9/2,-7/2,-5/2,-3/2}, plus one delta-m_F=2 pair with lower fidelity. Sec. IV explicitly states: 'when driving atoms in other states, we observed uncontrolled population transfers,' attributed to b ~ -3q, and the proposed remedy b > |q|(2F-1) is not implemented. The condition is concrete: for delta-m_F=1 Raman transitions, the resonance frequency is omega_m = b + q(2m+1). For q<0 and b>0, all these frequencies are positive and monotonically ordered only if b > |q|(2F-1). In the reported conditions, b/h = 960(5) Hz and q/h = -320 Hz, so b = -3q < |q|(2F-1) = 2560 Hz; the Zeeman spectrum is non-monotonic, with level crossings (e.g., at b = -3q, E_{-3/2} = E_{9/2}), and energy-difference signs reverse across the manifold. The paper's control demonstrations are confined to the negative-m side where the ordering is favorable. Thus the su(10)/complete-control statements in the abstract and conclusion rest on an unverified extrapolation to a higher-field regime. This does not invalidate the four-state interferometry, which is internally consistent, but it makes the full-toolkit claim conditional on a regime that is asserted, not demonstrated.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports experimental demonstration of coherent Raman manipulation of nuclear-spin Zeeman states of ultracold 87Sr (F=9/2), using a tensor light shift U_TLS(m_F)=q m_F^2 to spectrally isolate transitions between selected pairs. It shows Rabi oscillations for Δm_F=1 between |−5/2⟩ and |−3/2⟩, Δm_F=2 oscillations between |−7/2⟩ and |−3/2⟩, and Ramsey fringes between |−7/2⟩ and |−5/2⟩ that retain full contrast for 3 s when the TLS beam is off. It then implements two four-state protocols: two parallel Ramsey interferometers that simultaneously determine q and b, and an ancillary-state mapping that provides simultaneous estimates of two non-commuting collective spin components of an ensemble of effective qubits. The paper claims that these operations implement generators of su(10) and constitute a toolkit for fully exploiting the 10-state manifold.","tokens_in":22966,"tokens_out":5804,"duration_ms":58607,"significance":"The direct observations, Rabi oscillations, Ramsey contrast, the phase relations (1) and (2), and the ancillary-state readout, are clearly presented with error bars and parameter sets; the phase relations φ1=T(4q−b)/ℏ and φ2=T(8q−b)/ℏ are parameter-free given the quadratic-shift spectrum. The 3-s coherence of a nuclear-spin superposition and the parallelism of two interferometers are of genuine interest for quantum sensing and for SU(N)-symmetric quantum simulation. However, the full-manifold su(10) claim is stronger than the demonstrated subset: only four of ten states are controlled, and the paper's own Sec. IV reports uncontrolled transfers outside this subset for the applied field b≈−3q. The response to the skeptic's concern therefore lands: the central four-state interferometry is internally consistent, but the full-toolkit conclusion rests on an unverified extrapolation.","major_comments":[{"comment":"The central su(10)-toolkit claim is not supported by the reported data. The paper states that \"when driving atoms in other states, we observed uncontrolled population transfers,\" attributing this to b≈−3q, and the proposed remedy b>|q|(2F−1) is not implemented. For the reported values b/h=960(5) Hz and q/h=−320 Hz, one has b≈−3q and b<|q|(2F−1)=2560 Hz, so the Zeeman spectrum is non-monotonic and resonances for different pairs overlap. Thus the demonstrations are confined to m_F∈{−9/2,−7/2,−5/2,−3/2}, and the conclusion that concatenating Δm_F=1,2 pulses implements all of su(10) is an extrapolation to an untested regime. Either demonstrate the higher-field regime or restrict the abstract and conclusion to the demonstrated four-state subset.","section":"Sec. IV (last paragraph) and Sec. VI"},{"comment":"The statement \"We demonstrated these on four out of ten levels, with high fidelity >0.99\" is inconsistent with the Δm_F=2 Rabi data in Sec. IV, where the 1/e damping time of 40(5) ms limits the π/2-pulse fidelity to approximately 0.90. The >0.99 fidelity derived from the Δm_F=1 Rabi experiment cannot be transferred to the four-state demonstration as a whole. Please report the fidelity separately for each demonstrated transition.","section":"Sec. VI"}],"minor_comments":[{"comment":"The procedure of recalibrating η(−3/2) by up to 6% on the basis of extremal population estimates should be propagated into the error budget; otherwise the fidelity and population-normalization claims are not fully reproducible.","section":"Sec. III"},{"comment":"The green fit is a heuristic damped sine, and the red dashed curve comes from a master-equation simulation with an adjustable coherence-decay rate Γ_q; the text should state clearly that Γ_q is an empirical parameter fitted to these data and not independently measured.","section":"Sec. IV, Fig. 2a"},{"comment":"The derivation of Var(Ô_{y,z}) = Var(ŝ_{y,z}) + N_at/4 is stated without proof; a brief sketch or a pointer to supplemental material would make the statistical claim easier to verify.","section":"Sec. V.B, Eq. (7)"},{"comment":"The phrase \"simultaneously measures multiple observables ... including non-commuting ones\" could be misread; because the measured operators Ôz and Ôy commute (Eq. (6)), please state earlier that the scheme infers non-commuting observables of the original state from commuting measurements in an enlarged Hilbert space.","section":"Abstract and Sec. V.B"},{"comment":"The notation \"su(N)\" and \"su(2)\" appears in several places due to formatting; this should be corrected to \"SU(N)\" and \"SU(2)\" for consistency with standard usage.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The four-state experiments appear sound and the paper is a good fit for the journal. My main concern is that the abstract and conclusion overstate full-su(10) control beyond the demonstrated four-state subset; this is correctable either by adding the proposed higher-field test or by softening the claims. The fidelity inconsistency in Sec. VI should also be fixed. The stress-test concern about the untested b>|q|(2F−1) regime is, on reading the manuscript, valid and should be addressed head-on."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a genuinely useful experimental paper and it deserves serious refereeing. The new thing is that a tensor light shift can isolate Raman transitions between selected pairs of nuclear Zeeman states in 87Sr, so you can drive rotations that look like su(N) pair generators rather than just spin precession. The data support it: clean Rabi oscillations between mF=-5/2 and -3/2 with less than 1% leakage, and Ramsey fringes with no visible contrast loss over 3 s once the shift beam is off. The two parallel Ramsey interferometers for b and q, and the four-state scheme that reads out two non-commuting collective observables at once, are both new in this platform and convincingly demonstrated. Error bars and parameter sets are given, and the authors are mostly upfront about their limitations.\n\nThe soft spots are real but do not break the core. The abstract and conclusion promise full SU(10) control, but the experiments address only four of the ten states. In Sec. IV the authors state that driving other states produced uncontrolled population transfers, because b is close to -3q, and the proposed fix b > |q|(2F-1) is never implemented. So the complete-control claim is an extrapolation, not a result. That is the main issue. Second, the decoherence model uses an empirical Gamma_q that is fitted to the same Rabi data. It reproduces the curve but is not predictive. Third, detection efficiencies are recalibrated by up to 6% in places; this is minor because the central claims rest on phases and contrasts, not absolute populations. The phase relations phi1 and phi2 are parameter-free given the assumed spectrum, so the circularity burden is low.\n\nThe stress-test concern holds up: the four-state demonstrations are solid, but the full-manifold language needs adjustment or a supporting measurement at higher bias field. I would accept this for peer review, and I would ask the authors to either show the b > |q|(2F-1) regime or trim the su(10) claims. If they do that, this is a cite-worthy contribution to cold-atom qudit control and large-spin sensing.","headline":"Solid four-state demonstration of selective Raman rotations in 87Sr; the full-SU(10) control claim outruns the data.","tokens_in":23532,"tokens_out":2331,"would_cite":true,"duration_ms":22753,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A quadratic laser shift lets any chosen pair of strontium-87's ten nuclear spin states be rotated coherently while the other eight states stay untouched.","keywords":["tensor light shift","nuclear spin qudit","strontium-87","Raman transitions","SU(N) symmetry","Ramsey interferometry","simultaneous measurement","ultracold atoms"],"falsifier":"Drive a $\\delta m_F=1$ Raman transition on a pair not among $\\{-9/2,-7/2,-5/2,-3/2\\}$, such as $-1/2 \\leftrightarrow +1/2$, at the current bias $b\\simeq -3q$. If neighboring Zeeman populations change by more than a few percent over a $\\pi/2$ pulse, or if clean Rabi oscillations cannot be produced without multi-level leakage, the claim of full-manifold SU(10) control is falsified; the paper's own observation of uncontrolled transfers already indicates this failure mode, and the proposed fix of a higher field bias remains untested.","tokens_in":22311,"feed_emoji":"⚛️","tokens_out":5462,"duration_ms":52857,"temperature":0.7,"pith_summary":"The paper shows that a purely quadratic laser-induced shift of the ten Zeeman levels of fermionic strontium-87 makes Raman transitions between any chosen pair of spin states resonant while all other transitions stay far off resonance. The resulting evolution is a rotation generated by one SU(10) generator, not just precession of a spin vector. The authors demonstrate Rabi oscillations between a selected pair with no measurable population leakage, Ramsey coherence persisting beyond three seconds when the shift beam is off, and two four-state interferometer schemes: one sensing two external fields in parallel, the other reading out two non-commuting collective observables in a single shot.","feed_headline":"Laser shift selects pair rotations inside strontium's ten spin states","feed_subtitle":"Coherent pair rotations, 3-second Ramsey fringes, and parallel field sensing all follow from one quadratic energy shift.","key_machinery":"The central object is the tensor light shift $U_{TLS}(m_F)=q m_F^2$, produced by a $\\pi$-polarized beam tuned within the hyperfine structure of the $^1S_0 \\to {}^3P_1$ transition. Because adjacent Raman resonances are separated by multiples of $2q/h$, setting the Raman detuning picks out one pair of Zeeman states; with Raman intensity far below the tensor-shift beam intensity, off-resonant couplings are negligible. The resulting evolution is a rotation on a sub-Bloch sphere with generator $\\sigma^x_{m_F,m_F'}$, and concatenating such rotations with $\\sigma^z$ phases and coherent transfers to ancillary states produces the interferometric sequences that implement parallel field sensing and simultaneous non-commuting observable readout.","core_discovery":"The central claim is that a tensor light shift $U_{TLS}(m_F)=q m_F^2$ turns the ground-state $F=9/2$ manifold of $^{87}$Sr into a controllable qudit register: with the two-photon Raman coupling weak compared to $2|q|$, the dynamics between a selected pair $(m_F,m_F')$ is well approximated by $\\exp(-i\\theta \\sigma^x_{m_F,m_F'}/2)$, and the orthogonal manifold is essentially untouched. This implements unitary operations deriving from generators of the $\\mathrm{su}(10)$ algebra, beyond the spin-$F$ representation of $\\mathrm{su}(2)$. Concretely, the paper demonstrates Rabi oscillations between $m_F=-5/2$ and $-3/2$ with short-time contrast consistent with 1 and less than 1% population growth in the neighboring state, a Ramsey interferometer between $-7/2$ and $-5/2$ with no discernible contrast loss over 3 s when the tensor-shift beam is off, two parallel Ramsey interferometers that yield simultaneous readouts of the quadratic and linear Zeeman shifts, and a four-state measurement sequence that estimates two orthogonal collective pseudo-spin projections in a single experimental realization. The demonstrations use four of the ten states; the authors argue that the remaining pairs become addressable with a larger magnetic field bias $b>|q|(2F-1)$.","pith_inferences":["The same tensor-shift addressing should transfer to other fermionic alkaline-earth-like atoms, such as Yb, Cd, or Hg, whose $^3P_1$ hyperfine-to-linewidth ratio is favorable, potentially giving SU(N) control in species with larger nuclear spin.","Shaped Raman pulses or optimal control, as the paper's own simulations suggest, could operate at $2|q|/\\hbar\\Omega_R \\lesssim 1$ and perform full $\\mathrm{su}(10)$ unitaries in roughly $30\\hbar/|q|$, a speedup relative to the spectrally isolated regime used here.","The four-state non-commuting readout could be extended to all three spin projections of an ensemble of pseudo-spins 1/2, or to pseudo-spins larger than 1/2, at the cost of increased quantum projection noise, since ten states are available.","If the condition $b>|q|(2F-1)$ is not satisfied, the claimed complete SU(10) control over the whole manifold fails; the paper's data already show uncontrolled transfers outside the tested subset, so a direct test with other Zeeman pairs would settle the scope of the toolkit."],"forward_implications":["Rotations restricted to any pair of Zeeman states realize generators of $\\mathrm{su}(10)$, so arbitrary unitary operations and full spin-state tomography become possible in principle.","Long-lived Ramsey coherence, exceeding 3 s with the tensor-shift beam off, makes the nuclear-spin qubit useful for long-interrogation-time sensing and quantum information storage.","Parallel Ramsey interferometers measure $q$ and $b$ in each shot, enabling common-mode noise rejection and correlation analysis of field fluctuations.","The four-state measurement sequence gives simultaneous estimates of two orthogonal collective pseudo-spin projections, with only an added $N_{\\mathrm{at}}/4$ variance term.","Combining $\\delta m_F=1$ and $\\delta m_F=2$ transitions covers 17 generators, and with a larger bias $b>|q|(2F-1)$ the whole ten-state manifold should become addressable."],"supporting_citations":[{"why":"Supplies the tensor light shift mechanism used to separate Raman resonance conditions.","marker":"[31]"},{"why":"The optimal-control proposal for ten-level nuclear spin qudits in 87Sr that this experiment partially realizes.","marker":"[37]"},{"why":"Demonstrates generalized Ramsey interferometry with a single nuclear spin qudit, the analogue this work extends to atoms.","marker":"[15]"},{"why":"Provides the scheme for simultaneous readout of noncommuting collective spin observables, adapted here to strontium's ten-state manifold.","marker":"[45]"},{"why":"Supplies the spin-polarized state preparation and spin-selective detection used in all sequences.","marker":"[29]"},{"why":"Earlier coherent control of nuclear spin qubits limited to edge states, which tensor-light-shift addressing goes beyond.","marker":"[30]"},{"why":"Derives the effective light-induced Hamiltonian and the favorable hyperfine-to-linewidth ratio for low-loss tensor shifts.","marker":"[38]"},{"why":"Prior demonstration of non-Abelian geometric transformations in cold strontium, motivating control beyond simple spin precession.","marker":"[28]"}],"fun_headline_variants":["Tensor light shift turns strontium's 10 nuclear spins into a controllable qudit","Coherent control of all 10 nuclear spin states in strontium-87 with light","Raman rotations exploit tensor shift for 3-second coherent spin pairs","SU(10) engineering: strontium nuclear spin as a qudit resource","Parallel sensing across four nuclear spin states in strontium-87"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The toolkit's reach over all ten states assumes the tensor shift is purely quadratic and the linear shift $b$ is controllable enough that every Zeeman pair is spectrally isolated; the paper demonstrates this only for four states and reports uncontrolled transfers on other pairs because $b\\simeq -3q$ creates quasi-degeneracies.","fun_headline_variants_meta":{"raw":{"variants":["Tensor light shift turns strontium's 10 nuclear spins into a controllable qudit","Coherent control of all 10 nuclear spin states in strontium-87 with light","Raman rotations exploit tensor shift for 3-second coherent spin pairs","SU(10) engineering: strontium nuclear spin as a qudit resource","Parallel sensing across four nuclear spin states in strontium-87"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000495,"raw_usage":{"total_tokens":2500,"prompt_tokens":1090,"completion_tokens":1410,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":706,"completion_tokens_details":{"reasoning_tokens":1307}},"tokens_in":706,"tokens_out":1410,"duration_ms":11064,"temperature":1.0,"reasoning_tokens":1307,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:21:42.155167+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Drive a $\\delta m_F=1$ Raman transition on a pair not among $\\{-9/2,-7/2,-5/2,-3/2\\}$, such as $-1/2 \\leftrightarrow +1/2$, at the current bias $b\\simeq -3q$. If neighboring Zeeman populations change by more than a few percent over a $\\pi/2$ pulse, or if clean Rabi oscillations cannot be produced without multi-level leakage, the claim of full-manifold SU(10) control is falsified; the paper's own observation of uncontrolled transfers already indicates this failure mode, and the proposed fix of a higher field bias remains untested.","supporting_citations":[{"cited_title":"Deutsch and P","cited_arxiv_id":null,"evidence_quote":"Supplies the tensor light shift mechanism used to separate Raman resonance conditions."},{"cited_title":"Godfrin, R","cited_arxiv_id":null,"evidence_quote":"Demonstrates generalized Ramsey interferometry with a single nuclear spin qudit, the analogue this work extends to atoms."},{"cited_title":"Bataille, A","cited_arxiv_id":null,"evidence_quote":"Supplies the spin-polarized state preparation and spin-selective detection used in all sequences."},{"cited_title":"Barnes, P","cited_arxiv_id":null,"evidence_quote":"Earlier coherent control of nuclear spin qubits limited to edge states, which tensor-light-shift addressing goes beyond."},{"cited_title":"Burba, H","cited_arxiv_id":null,"evidence_quote":"Derives the effective light-induced Hamiltonian and the favorable hyperfine-to-linewidth ratio for low-loss tensor shifts."}],"review_version":1}