{"id":"6a858215-183b-4502-a797-dd53d4b62079","arxiv_id":"2412.08043","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"YbNi4Mg single crystals show a superheavy-fermion state with gamma0 = 5.65 J mol^-1 K^-2, Wilson ratio 32.1, short-range magnetic correlations below 0.3 K, and magnetocaloric cooling comparable to Gd3Ga5O12.","lead":"Researchers grew the first single crystals of the metallic compound YbNi4Mg and found that it becomes a 'superheavy fermion' material at very low temperatures, with an enormous electronic heat capacity and no magnetic ordering down to 0.07 K. The material cools itself strongly when its magnetic field is removed, performing comparably to a standard magnetic refrigerant used for sub-Kelvin cooling.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The superheavy-fermion claim rests on isolating γ0 by fitting (C−Cl)/T below 0.1 K to γ0+AnT^-3; if a short-range magnetic or distributed-hyperfine term enters that window, γ0 and hence RW=32.1 are not uniquely determined. A reanalysis with an unconstrained power law would settle it.","rationale":"The paper is a competent experimental work, and the superheavy-fermion phenomenology has multiple independent indicators: the large low-T C/T, the Kondo-like resistivity with a T^2 coefficient, the lack of ordering anomalies, and a magnetic entropy approaching Rln2. My stress test does not attack those facts. The single most load-bearing quantitative number is γ0, because it is used both to certify the SHF state and to compute RW. The reader's weakest-assumption analysis already identified the low-T fit; I agree and sharpen it: the fit window is very close to the T* maximum, and the assumed T^-3 nuclear form is not independently established, so a correlated γ0/An trade-off is possible. The proposed reanalysis with a free exponent is a direct computational check; if the raw data are unavailable, the reported γ0 should carry a larger systematic uncertainty. This does not warrant rejection, since the qualitative conclusions can survive a moderate change in γ0; it confirms the need for the conditional verdict and for raw-data transparency.","tokens_in":17986,"tokens_out":14387,"duration_ms":163520,"concrete_test":"Reanalyze the raw zero-field (C−Cl)/T data below 0.1 K with the model γ0 + AnT^-α, letting α be free, and repeat with the fit upper boundary set to 70, 80, and 100 mK; report γ0, An, α, and their covariance. If the best-fit α is not 3, or if γ0 varies by more than about 20% across the three windows, then the zero-field γ0 = 5.65 J mol^-1 K^-2 is a fitting artifact rather than an isolated electronic term, and RW = 32.1 should be revised accordingly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section III and Fig. 5(b) inset: the zero-field value γ0 = 5.65 J mol^-1 K^-2 is obtained from a two-parameter fit of (C−Cl)/T below 0.1 K to γ0 + AnT^-3, where the An term is treated as the high-temperature tail of a single nuclear Schottky anomaly. This number directly feeds the superheavy-fermion classification, the Wilson ratio RW = 32.1, and the Kadowaki-Woods ratio; it is therefore the most load-bearing step in the paper. The fit assumes that the only temperature-dependent contribution in the window is one nuclear term and that the electronic term is already T-independent. Those assumptions are not tested. The same data show C/T ∝ T^-2 above T* ≈ 0.3 K and a broad maximum at T*, so a short-range magnetic contribution with a different power law could persist below 0.1 K and be absorbed into either γ0 or An. With a fit window of only a factor of two in temperature and no reported error bars or raw data, the decomposition is not unique. A separate, smaller issue is that χ0 = 2.49 emu/mol is taken at 0.4 K and 0.1 T, so RW is built from two finite-temperature proxies. The qualitative SHF phenomenology is well supported; the quantitative value of γ0 is the fragile part.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports the synthesis of single-crystalline YbNi4Mg and a low-temperature study combining resistivity, magnetization, specific heat down to below 0.1 K, and quasi-adiabatic demagnetization measurements. It claims a superheavy-fermion state with electronic specific-heat coefficient γ0 = 5.65 J mol^-1 K^-2, an elevated Wilson ratio RW = 32.1, a Kondo temperature TK ≈ 0.9 K, and no long-range magnetic order down to 70 mK. A broad specific-heat maximum at T* ≈ 0.3 K is attributed to short-range cooperative magnetism, and the magnetocaloric response is shown to cool a sample from 2 K to 0.21 K starting from 8 T, outperforming GGG under comparable starting conditions.","tokens_in":18314,"tokens_out":4010,"duration_ms":42772,"significance":"If the quantitative characterization holds, this paper is a valuable addition to the small family of superheavy-fermion compounds and strengthens the case for using such materials in sub-kelvin adiabatic demagnetization refrigeration. The qualitative phenomenology is convincingly supported by multiple independent probes: the T^2 resistivity, the magnetoresistivity scaling with T0 = 0.25 K, the absence of a sharp ordering anomaly in C(T), the field-dependent Weiss temperature, and the direct T(B) traces all consistently point to a weak-coupling Kondo lattice with short-range correlations. The direct comparison of the cooling performance with GGG in Fig. 9 is a useful practical benchmark. The main weakness is that the headline quantitative claims — γ0, RW, and all derived ratios — rest on a two-component decomposition of the specific heat in a narrow temperature window and on a finite-temperature value of χ0, neither of which is accompanied by an uncertainty estimate or a test of the assumed functional form.","major_comments":[{"comment":"The central value γ0 = 5.65 J mol^-1 K^-2 is obtained by fitting (C−Cl)/T below 0.1 K to γ0 + An T^-3, assuming that the only temperature-dependent contribution in that window is the high-temperature tail of a single nuclear Schottky term and that the electronic term is already T-independent. This assumption is not tested: the same data show C/T ∝ T^-2 above T* ≈ 0.3 K and a broad maximum at T*, so a short-range magnetic contribution with a different power law could persist below 0.1 K and be absorbed into either γ0 or An. The authors should fit (C−Cl)/T to γ0 + An T^-α with α free, estimate An independently from the known 171Yb/173Yb hyperfine parameters, and report the fit residuals and parameter uncertainties. Without such a test, the superheavy-fermion classification, RW = 32.1, and RKW all inherit an unquantified systematic error.","section":"§III, Fig. 5(b) inset"},{"comment":"The Wilson ratio RW = 32.1 is computed using χ0 = 2.49 emu/mol measured at T = 0.4 K and B = 0.1 T, not at the zero-temperature limit. Since χ(T) rises steeply on cooling and short-range correlations develop below 0.3 K, the T→0 value of χ0 could differ substantially from the 0.4 K value. The authors should either provide a χ(T) extrapolation to T→0 or quote RW with a stated uncertainty that includes this finite-temperature ambiguity. In addition, the entries in Table I for other SHF compounds use χ values measured at different temperatures and fields, so the cross-material claim that large RW is a common feature of SHF compounds without long-range order is not yet quantitatively robust.","section":"§IV, Wilson ratio"},{"comment":"The statement that no magnetic ordering is observed down to 70 mK is based primarily on the absence of a sharp anomaly in C(T), on ρ(B) data at 0.1 K, and on χ(T) data that extend only to 0.4 K. The authors should state this limitation explicitly and, if possible, provide a low-field χ measurement below 0.4 K or μSR data to support the absence of static order. As written, the combination of the T^-2 rise in C/T above T* and the low-temperature upturn could also be consistent with a distribution of hyperfine fields or a frozen spin state, which would directly affect the γ0 decomposition and the interpretation of the ground state.","section":"§III, Fig. 5(a) and Fig. 8"}],"minor_comments":[{"comment":"The title in the manuscript text reads 'magne tocaloric effect' with a spurious space, and the introduction contains 'miliKevin' instead of 'millikelvin'; both should be corrected.","section":"Title and introduction"},{"comment":"References [38] and [39] are the same Desgranges–Schotte paper and should be merged into a single citation.","section":"References"},{"comment":"The caption contains the typo 'Schottcky' for 'Schottky' in two places.","section":"Fig. 5 caption"},{"comment":"The text refers to a device sketch as 'to be sketched in Fig. 9(a) inset' and later to 'Fig. 9(c) inset', but the figure has multiple insets; please clarify which panel contains the thermal-stage drawing and which contains the PPMS cooling-device photograph.","section":"Fig. 9"},{"comment":"The footnotes for YbPt2Sn and Ce4Pt12Sn25 give the temperatures at which χ0 is read but not the magnetic field values; providing the field would make the Wilson-ratio comparison reproducible.","section":"Table I"}],"recommendation":"major_revision","confidential_remarks":"This is a well-executed experimental characterization with a compelling qualitative story, but the two most prominent quantitative results — γ0 = 5.65 J mol^-1 K^-2 and RW = 32.1 — are not yet sufficiently secured. The stress-test concern about the specific-heat decomposition below 0.1 K is legitimate and should be addressed with additional fitting tests or independent hyperfine estimates before publication. The paper would also be stronger if the authors provided estimates of statistical and systematic uncertainties for the key parameters."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know about this one: it is a careful experimental study of YbNi4Mg single crystals, and the qualitative superheavy-fermion (SHF) claim is well supported. The paper deserves a serious referee, but the headline numbers—gamma0 = 5.65 J/mol K^2 and Wilson ratio RW = 32.1—are shakier than the text implies.\n\nWhat is genuinely new: first single-crystal growth, thermodynamic and transport data down to 70 mK, and demonstration that this metal cools magnetocalorically below 0.3 K from 2 K, beating GGG under comparable quasi-adiabatic conditions. The C/T shows the familiar SHF pattern: a T^-2 rise above T* ≈ 0.3 K, a broad maximum, no long-range order, and a nuclear Schottky upturn below 0.1 K. The entropy reaches R ln2 by about 4 K, and the field dependence of C/T and the magnetoresistivity scaling are consistent with a weak Kondo scale TK ≈ 0.9 K. The LuNi4Mg lattice subtraction and the CEF analysis are competently done.\n\nThe soft spots are the two most publicized numbers. gamma0 is extracted from a two-parameter fit of (C−Cl)/T below 0.1 K to γ0 + An T^-3. The fit window is only about a factor of two in temperature, no error bars are reported, and the decomposition assumes the only non-electronic contribution below 0.1 K is a single nuclear Schottky term. A distributed hyperfine field or a residual short-range magnetic term with a different power law would shift gamma0, and therefore RW. This is not a fatal flaw for the SHF assignment—even a gamma0 of 3 J/mol K^2 would still be superheavy—but the precise value and the KW ratio should be treated as provisional until a reanalysis with an unconstrained power law, or raw data, is provided. Second, χ0 = 2.49 emu/mol is taken at 0.4 K and 0.1 T rather than extrapolated to T → 0; given the rise in χ below 1 K, RW = 32.1 is an overestimate, though likely still large.\n\nThere are minor presentation issues (duplicate reference for Desgranges–Schotte, a few typos) that don't affect the science.\n\nFor a referee report, ask for the raw specific-heat data below 0.2 K in zero and applied fields, a fit allowing a free power law for the low-T upturn, and a clearer statement of how χ0 is defined. If those are addressed, the paper is a solid addition to the SHF literature and to the sub-Kelvin cooling discussion.\n\nI'd send this to peer review, and I'd read the revised version carefully.","headline":"A solid new single-crystal superheavy-fermion candidate whose qualitative claim holds, but the headline gamma0 and Wilson ratio rest on a fit that is more fragile than the abstract suggests; it still deserves peer review.","tokens_in":18938,"tokens_out":3370,"would_cite":true,"duration_ms":33076,"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":"YbNi4Mg is a superheavy-fermion metal with γ0 = 5.65 J mol^-1 K^-2 and Wilson ratio RW = 32.1, and demagnetization from 8 T cools it from 2 K to 0.21 K, rivaling GGG.","keywords":["superheavy fermion","heavy fermion","YbNi4Mg","Wilson ratio","magnetocaloric effect","adiabatic demagnetization refrigeration","short-range magnetic order","Kondo lattice"],"falsifier":"Measure $C/T$ below 0.1 K at several applied fields and check that the fitted constant electronic term $\\gamma_0$ stays unchanged while the nuclear $T^{-3}$ term changes with field as expected; if $\\gamma_0$ shifts when the fit window is varied or the decomposition fails, the central claim collapses. Alternatively, measure the susceptibility below 0.4 K in 0.1 T: if its zero-temperature limit is not close to 2.49 emu/mol, the Wilson ratio of 32.1 is not a zero-temperature statement.","tokens_in":17746,"feed_emoji":"🧲","tokens_out":16344,"duration_ms":143146,"temperature":0.7,"pith_summary":"This paper aims to establish that the intermetallic compound YbNi$_4$Mg is a superheavy-fermion metal, a state in which the electrons behave as if their effective mass is thousands of times larger than in an ordinary metal, and that this state coexists with short-range cooperative magnetism rather than long-range magnetic order down to 70 mK. The central evidence is a very large electronic specific-heat coefficient $\\gamma_0 = 5.65$ J mol$^{-1}$ K$^{-2}$ and a Wilson ratio $R_W = 32.1$, far above the value near 2 expected for a conventional heavy-fermion Fermi liquid. The authors argue that this enhanced Wilson ratio, seen also in other superheavy-fermion compounds without magnetic order, is a fingerprint of residual spin fluctuations, likely of a spin-liquid-like character. They then show that the same field-sensitive heavy-electron state produces a strong magnetocaloric effect: quasi-adiabatic demagnetization from 8 T and 2 K cools a 160 mg sample to 0.21 K, below what the standard refrigerant Gd$_3$Ga$_5$O$_{12}$ reaches under similar conditions.","feed_headline":"Removing an 8-T field cools YbNi4Mg to 0.21 K","feed_subtitle":"The intermetallic's Wilson ratio of 32.1 hints at spin-liquid-like fluctuations behind the cooling.","key_machinery":"The load-bearing object is the superheavy-fermion state, quantified by the zero-temperature electronic specific-heat coefficient $\\gamma_0$ and by the Wilson ratio $R_W = (\\pi^2 k_B^2 / 3 \\mu_B^2)\\,\\chi_0/\\gamma_0$, which compares spin-susceptibility enhancement with specific-heat enhancement. The paper isolates $\\gamma_0$ by subtracting the lattice heat capacity of LuNi$_4$Mg and fitting the $C/T$ rise below 0.1 K to $\\gamma_0 + A_n T^{-3}$, where the $T^{-3}$ term is the high-temperature tail of the nuclear magnetic contribution from Yb nuclei in the hyperfine field of the $4f$ electrons; the same magnetic entropy, fed through a spin-1/2 Kondo-model relation, gives $T_K \\approx 0.9$ K. The argument is carried by three synchronized signatures: the $C/T$ maximum at $T^* \\approx 0.3$ K, the shoulders in $d\\chi/dT$ and $d\\rho/dT$, and the field-induced crossover near $B^* \\approx 0.1$ T seen in $C(B)$, magnetoresistance, and the $T(B)$ demagnetization trace. Together they tie the very large $\\gamma_0$ to short-range cooperative magnetism and make the entropy release field-sensitive enough to produce the observed sub-Kelvin cooling.","core_discovery":"YbNi$_4$Mg realizes a superheavy-fermion ground state: after subtracting the lattice contribution of LuNi$_4$Mg and a low-temperature nuclear term, the electronic specific-heat coefficient is $\\gamma_0 = 5.65$ J mol$^{-1}$ K$^{-2}$, and the low-field susceptibility gives a Wilson ratio $R_W = 32.1$. No long-range magnetic order appears down to 70 mK; instead a broad $C/T$ maximum at $T^* = 0.3$ K, with matching shoulders in $d\\chi/dT$ and $d\\rho/dT$, marks short-range cooperative magnetism entwined with the heavy-electron liquid. A field of about 0.1 T drives a metamagnetic-like crossover, and quasi-adiabatic demagnetization from 8 T and 2 K reaches 0.21 K (from 3 T, 0.42 K), cooling below Gd$_3$Ga$_5$O$_{12}$ under comparable conditions. The paper takes the large Wilson ratio and the reduced Kadowaki-Woods ratio as evidence that residual spin fluctuations, possibly spin-liquid-like, distinguish the non-ordering superheavy-fermion state from a conventional heavy-fermion Fermi liquid.","pith_inferences":["A direct check of the spin-liquid interpretation would be muon spin rotation or neutron scattering: if the Wilson ratio reflects residual spin dynamics, YbNi$_4$Mg should show persistent dynamic spin fluctuations without static order below $T^*$, with the fluctuation spectrum softening near $B^* \\approx 0.1$ T.","The refrigerant claim could be sharpened into an engineering comparison by measuring thermal conductivity and cooling power per gram against Gd$_3$Ga$_5$O$_{12}$ and the classic hydrated paramagnetic salts; the paper demonstrates the effect but does not give a full refrigeration-cycle comparison.","Because the nuclear term comes from the magnetic moments of Yb nuclei, changing the Yb isotope mix would shift $A_n$ without changing the electronic $\\gamma_0$, providing an experimental test of the specific-heat decomposition.","The comparative pattern in the paper implies a testable trend: if long-range order is introduced into a related superheavy-fermion compound by pressure or chemical substitution, the Wilson ratio should fall from values near 32 toward the Fermi-liquid value near 2 while the broad $C/T$ maximum sharpens into a transition."],"forward_implications":["If the compound is a superheavy fermion, its $\\gamma_0 = 5.65$ J mol$^{-1}$ K$^{-2}$ and $A = 22.17$ $\\mu\\Omega$ cm K$^{-2}$ place it in the same class as YbCu$_4$Ni and YbPt$_2$Sn, and the reduced Kadowaki-Woods ratio shows that its thermodynamic mass enhancement exceeds its transport enhancement.","The absence of long-range order down to 70 mK, together with a broad $C/T$ maximum at $T^* \\approx 0.3$ K, means the magnetic degrees of freedom remain cooperative but short-ranged, entangled with the heavy-electron state rather than frozen into an ordered lattice.","A field of roughly 0.1 T suppresses the short-range order and the heavy-electron state, producing a metamagnetic-like crossover at which the magnetic Grüneisen ratio changes sign without diverging, so the crossover is not a quantum critical point.","Quasi-adiabatic demagnetization from 8 T and 2 K reaches 0.21 K (from 3 T, 0.42 K) with a 160 mg sample, and under similar initial conditions YbNi$_4$Mg cools below Gd$_3$Ga$_5$O$_{12}$, supporting superheavy-fermion metals as sub-Kelvin refrigerants.","The large Wilson ratio $R_W = 32.1$, together with similar values in other non-ordering superheavy-fermion compounds and ordinary values in ordering ones, supports residual spin fluctuations as a defining feature of the non-ordering superheavy-fermion ground state."],"supporting_citations":[{"why":"It establishes YbCu4Ni as the benchmark superheavy-fermion compound and provides the γ0 ≈ 7.5 J mol^-1 K^-2 comparison value used in this paper.","marker":"[3]"},{"why":"It supplies the precedent of a temperature-field minimum at B* ≈ 0.1 T in a superheavy-fermion compound, the direct magnetocaloric comparison.","marker":"[10]"},{"why":"It provides the YbCo2Zn20 metamagnetic-crossover and Grüneisen-ratio behavior used to interpret the field dependence and absence of divergence.","marker":"[11]"},{"why":"It gives the earlier polycrystalline YbNi4Mg susceptibility data that this work extends with single crystals to much lower temperatures.","marker":"[16]"},{"why":"It documents the C/T ~ T^-2 short-range-correlation behavior and broad low-temperature maxima in YbPt2Sn and YbPt2In.","marker":"[19]"},{"why":"It describes the quasi-adiabatic thermal device and shielding geometry used for the larger-mass demagnetization cooling demonstration.","marker":"[33]"},{"why":"It gives the scaling form B/(T+T0) used to collapse the magnetoresistivity and extract the weak Kondo coupling T0 = 0.25 K.","marker":"[34]"},{"why":"It provides the spin-1/2 Kondo-model entropy relation used to convert the measured magnetic entropy into TK ≈ 0.9 K.","marker":"[38]"},{"why":"It supplies the theoretical result that a large Wilson ratio can arise from spin-liquid-like fluctuations in a frustrated spin-orbital system, supporting the paper's interpretation.","marker":"[49]"}],"fun_headline_variants":["Superheavy fermion YbNi4Mg demagnetizes to 0.21 K","YbNi4Mg: superheavy fermion, Wilson ratio 32, cools to 0.21 K","Field removal cools YbNi4Mg to 0.21 K, rivaling Gd3Ga5O12","Superheavy fermion YbNi4Mg shows magnetocaloric effect, cools to 0.21 K"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The headline electronic specific heat rests on the assumption that everything in the measured $C/T$ below 0.1 K is either a constant electronic term or a nuclear term from the magnetic moments of Yb nuclei, with no other contribution; if a different source of heat capacity enters that range, $\\gamma_0 = 5.65$ J mol$^{-1}$ K$^{-2}$ and the Wilson ratio built on it would change.","fun_headline_variants_meta":{"raw":{"variants":["Superheavy fermion YbNi4Mg demagnetizes to 0.21 K","YbNi4Mg: superheavy fermion, Wilson ratio 32, cools to 0.21 K","Field removal cools YbNi4Mg to 0.21 K, rivaling Gd3Ga5O12","Superheavy fermion YbNi4Mg shows magnetocaloric effect, cools to 0.21 K"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001058,"raw_usage":{"total_tokens":4503,"prompt_tokens":1074,"completion_tokens":3429,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":690,"completion_tokens_details":{"reasoning_tokens":3320}},"tokens_in":690,"tokens_out":3429,"duration_ms":27212,"temperature":1.0,"reasoning_tokens":3320,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T18:16:31.624604+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $C/T$ below 0.1 K at several applied fields and check that the fitted constant electronic term $\\gamma_0$ stays unchanged while the nuclear $T^{-3}$ term changes with field as expected; if $\\gamma_0$ shifts when the fit window is varied or the decomposition fails, the central claim collapses. Alternatively, measure the susceptibility below 0.4 K in 0.1 T: if its zero-temperature limit is not close to 2.49 emu/mol, the Wilson ratio of 32.1 is not a zero-temperature statement.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It establishes YbCu4Ni as the benchmark superheavy-fermion compound and provides the γ0 ≈ 7.5 J mol^-1 K^-2 comparison value used in this paper."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It supplies the precedent of a temperature-field minimum at B* ≈ 0.1 T in a superheavy-fermion compound, the direct magnetocaloric comparison."},{"cited_title":"Tokiwa, B","cited_arxiv_id":null,"evidence_quote":"It provides the YbCo2Zn20 metamagnetic-crossover and Grüneisen-ratio behavior used to interpret the field dependence and absence of divergence."},{"cited_title":"Linsinger, M","cited_arxiv_id":null,"evidence_quote":"It gives the earlier polycrystalline YbNi4Mg susceptibility data that this work extends with single crystals to much lower temperatures."},{"cited_title":"Gruner, D","cited_arxiv_id":null,"evidence_quote":"It documents the C/T ~ T^-2 short-range-correlation behavior and broad low-temperature maxima in YbPt2Sn and YbPt2In."},{"cited_title":"Xiang, C","cited_arxiv_id":null,"evidence_quote":"It describes the quasi-adiabatic thermal device and shielding geometry used for the larger-mass demagnetization cooling demonstration."},{"cited_title":"SCHLOTTMANN, ZEITSCHRIFT FUR PHYSIK B-CONDENSED MATTER 51, 223 (1983)","cited_arxiv_id":null,"evidence_quote":"It gives the scaling form B/(T+T0) used to collapse the magnetoresistivity and extract the weak Kondo coupling T0 = 0.25 K."},{"cited_title":"Chen and Y","cited_arxiv_id":null,"evidence_quote":"It supplies the theoretical result that a large Wilson ratio can arise from spin-liquid-like fluctuations in a frustrated spin-orbital system, supporting the paper's interpretation."}],"review_version":1}