{"id":"e96541e7-851c-4fff-9468-2f10bcfb7a3c","arxiv_id":"hep-th/9601029","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":9.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"The statistical degeneracy of D-brane bound states in five-dimensional string theory matches the Bekenstein-Hawking entropy of the corresponding extremal black hole at leading order in large charge.","lead":"By counting the quantum states of D-branes wrapped on cycles of a compact surface, the authors reproduce the thermodynamic entropy of a five-dimensional extremal black hole, including the factor of one quarter in the area law. It is the founding calculation of black hole microstate counting and the template for the program that followed.","discovery_kind":"first_principles","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"The leading-order match between Eqs. (3.4) and (2.11) is genuine, but the two sides are derived in different asymptotic regimes; the paper never states the joint condition Q_H >> Q_F^2 >> 1 under which both expansions are controlled.","rationale":"The paper does what it claims: two independent computations, neither using the other, agree at leading order in large charge. The gravity side (Eqs. 2.8–2.11) is a straightforward extremal Reissner-Nordström-type solution; the brane side (Eqs. 3.1–3.4) is a Cardy count on a hyperkähler sigma model. The target A/4 is not an input to the count, so the agreement is non-circular. I give the paper credit for its candor: it states the match is leading order, that the sigma-model identification was conjectured in [29] and checked in [30,31] only for restricted charges, that the elliptic genus bounds rather than equals the degeneracy (with [38] cited for the leading-order equivalence), and in footnote 8 that large-c corrections to Cardy are not understood. The reader's weakest_assumption — the symmetric-product identification — is a real soft spot but is partially defused by footnote 7, which notes that only dim M = 4k enters the exponent, and c = 6k follows from hyperkählerity rather than from the detailed sigma model. That is why I mark agreement as partial rather than agree. The sharper concern is the regime mismatch: the count is stated for large Q_H at fixed Q_F, the gravity result for both charges large, and the Cardy formula needs n >> c, so the comparison requires Q_H >> Q_F² >> 1. The paper never writes this joint condition, and the abstract's unqualified 'is derived' [abstract:s0] reads stronger than the body's overlap-of-asymptotics argument. This is a framing and precision issue, not an error in the equations, and it does not move the verdict: the leading terms do agree in the overlap, and the result has been extended widely since. I would keep ACCEPT and ask only that the joint regime be stated explicitly in any revision. No significant objection lands against the central claim itself.","tokens_in":9374,"tokens_out":6959,"duration_ms":99820,"concrete_test":"Independently compute the exact elliptic-genus Fourier coefficients of Sym^k(K3) for k = 2 and k = 3 (using the DVV product formula, which postdates this paper) at levels n = 10, 20, 50, and fit ln|coefficient| against 2π√(n·6k/6). If the fitted effective central charge differs from 6k by more than the predicted c/(12n) correction, then c = 6k is not what drives the growth and Eq. (3.4) is not established. In parallel, estimate the leading gravity-side corrections to (2.11) (string loop ~ Q_F/Q_H, α' ~ 1/√(Q_F² Q_H)) and check that in the joint regime Q_H >> Q_F² >> 1 they are parametrically smaller than the Cardy correction 3Q_F²/Q_H. If they are not, the two expansions have no common leading term and the match is accidental at leading order only.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is an asymptotic match: S_stat = 2π√(Q_H(Q_F²/2+1)) from Cardy growth of the elliptic genus of Sym^k(K3), k = Q_F²/2+1, versus S_BH = 2π√(Q_H Q_F²/2) from the horizon area of the extremal solution. The leading terms agree when Q_F²/2 >> 1. The soft spot is not the symmetric-product identification the reader flags: footnote 7 explicitly says only dim M = 4k matters, and c = 6k follows from hyperkählerity, so the precise sigma model is not load-bearing for the exponent. The soft spot is the regime of validity. Section 1 states the count for 'large Q_H and fixed Q_F' (Eq. 1.1); section 2 states that (2.11) is reliable only when both Q_H and Q_F are large. The Cardy formula (3.2) as applied needs n >> c, i.e. Q_H >> 3Q_F², not merely n >> 1. So the comparison lives in the overlap Q_H >> Q_F² >> 1, a joint condition the paper never writes down. Footnote 8 ('large c corrections ... to determine the range of validity') gestures at this but does not connect it to the gravity-side expansion. In the strict fixed-Q_F limit where the count is clean, the gravity side has string-loop corrections of order Q_F/Q_H and α' corrections of order 1/√(Q_F² Q_H); in the large-Q_F limit where gravity is clean, the Cardy formula has 1/c corrections. The match is therefore an overlap-of-asymptotics argument, legitimate but weaker than the abstract's unqualified 'is derived' [abstract:s0]. This is a precision-of-framing concern, not an error: the leading terms do agree.","agreement_with_reader":"partial"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript derives the Bekenstein-Hawking entropy S_BH = A/4 for a class of five-dimensional extremal black holes in string theory by counting BPS bound states. Section 2 constructs the extremal dyonic solution of the five-dimensional N=4 supergravity obtained from type II on K3 x S1 (or heterotic on T^5), fixes the horizon dilaton through Eq. (2.6), and obtains the horizon area Eq. (2.10) and entropy Eq. (2.11), S_BH = 2 pi sqrt(Q_H Q_F^2/2). Section 3 identifies the corresponding BPS states with D-branes wrapped on supersymmetric cycles of K3 times S1, invokes the conjectured symmetric-product sigma model Eq. (3.1) with c = 6(Q_F^2/2 + 1), and applies the Cardy growth formula Eq. (3.2) at level n = Q_H to obtain S_stat = 2 pi sqrt(Q_H (Q_F^2/2 + 1)), Eq. (3.4). The two expressions agree to leading order for large charges, which is the paper's central claim. Section 4 discusses the strong-coupling regime of the D-brane description and implications for the information puzzle.","tokens_in":9717,"tokens_out":4114,"duration_ms":63660,"significance":"If the leading-order match holds, this is a genuine microscopic derivation of the area law for a controlled class of extremal black holes, and it reproduces the numerical coefficient, not merely the scaling. The strength of the argument is that the two sides are computed independently: the horizon area follows from the effective action and the solution Eqs. (2.8)-(2.9), while the state count follows from the elliptic genus of a hyperkahler sigma model, with no input of the target answer A/4. The result is parameter-free in the sense that no adjustable constant is fitted between Eqs. (2.11) and (3.4), and it yields a sharp, checkable prediction for the asymptotic degeneracy of BPS states at charges (Q_F, Q_H). The paper is also careful about what it does not claim: it flags the elliptic-genus-versus-degeneracy distinction, the conjecture status of Eq. (3.1), and the breakdown of the D-brane description at strong coupling. These are the marks of a substantial and honestly framed result.","major_comments":[{"comment":"The comparison of Eqs. (3.4) and (2.11) is an overlap-of-asymptotics argument, but the manuscript never states the joint regime in which both expansions are controlled, and the two sections describe the regime differently. Section 1 specifies the count for 'large Q_H and fixed Q_F' (Eq. 1.1), whereas Section 2 states that Eq. (2.11) is reliable only when both Q_H and Q_F are large, since string-loop corrections scale as Q_F/Q_H and sigma-model corrections as 1/sqrt(Q_F^2 Q_H). On the counting side, the Cardy formula Eq. (3.2) is applied with n = Q_H and c = 6(Q_F^2/2 + 1), and its asymptotic form requires n >> c, i.e. Q_H >> 3 Q_F^2, not merely n >> 1 as written below Eq. (3.2). The actual domain of common validity is therefore Q_H >> Q_F^2 >> 1, which is narrower than either section's stated condition and is nowhere written down. Footnote 8 gestures at large-c corrections but does not connect them to the gravity-side expansion. This does not affect the correctness of the leading-order agreement, but it does affect the framing: the abstract's unqualified statement that the relation 'is derived' [abstract:s0] is stronger than what the body establishes, and the paper's own Eq. (1.1) versus Eq. (1.2) differ by the additive one inside the square root. I recommend that the authors state the joint condition explicitly, most naturally as a sentence at the end of Section 3 and a corresponding qualification in the abstract, and that they note which corrections spoil the match at each edge of the overlap region.","section":"§1 Eq. (1.1); §2 Eq. (2.11); §3 Eqs. (3.2)-(3.4)","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The conjecture status of the symmetric-product identification is acknowledged in the text and in footnote 7, and footnote 7 correctly observes that only dim M = 4(Q_F^2/2 + 1) enters the exponent, so the precise sigma model is not load-bearing for the leading growth. Since this robustness is the reason the result survives the restricted verification of [30,31], it would be worth promoting that observation from a footnote into the main text, so that a reader who stops at Eq. (3.1) does not overestimate the dependence on the conjecture.","section":"§3, Eq. (3.1) and footnote 7","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The manuscript correctly notes that the elliptic genus is a signed sum and therefore bounds rather than equals the degeneracy, and cites [38] for the equality of leading asymptotics. A single sentence quantifying this, for instance that the two differ at most by a subexponential factor and hence not in ln d at leading order, would make the logic of the comparison self-contained.","section":"§3, paragraph following Eq. (3.1)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The identification Q_H = P proceeds through a IIB-to-IIA duality on the circle followed by the interpretation of momentum as winding charge. This step is load-bearing for the comparison, since it is what makes the counted states the same objects as the solution of Section 2, and it is dispatched in a few lines. Expanding it slightly, with the duality chain written out, would help readers who are not already fluent in the mapping.","section":"§3, charge identification below Eq. (3.1)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The normalization conventions that make Q_H and Q_F^2/2 integers are stated in footnote 4 and in the text around Eq. (2.4), but the numerical factors in Eqs. (2.4), (2.6) and (2.9) are easy to lose track of. A short summary sentence collecting the conventions (alpha' = G_N = 1, the charge normalizations, and the definition Q_F^2 = Q_R^2 - Q_L^2) at the point where the solution is written would reduce the burden on the reader.","section":"§2, Eqs. (2.3)-(2.6) and footnote 4","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The text defining the BPS condition uses L_0 and L-bar_0, but the distinction between left- and right-moving Virasoro generators is not typographically clear in the typeset paragraph, and the same symbol appears to be used for both. Since the whole counting argument rests on setting one chirality to its vacuum, the notation should be made unambiguous.","section":"§3, notation for left- and right-movers","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"There are numerous typographical errors that should be corrected: the author name appears as 'V afa' on the title page; 'Robinson-Berttoti' should read 'Robinson-Bertotti'; 'compnent' in the paragraph after Eq. (2.1); 'worlvolume' in Section 3; 'genu s' in footnote 9; 'dimensions of M grows' should be 'dimension of M grows' in footnote 7; 'a functions of Q_F^2 and Q_H' should be 'a function of' in footnote 3; 'Schwarzchild' should be 'Schwarzschild' in Section 2; and several words are broken by stray spaces ('co unting', 'b ehaves', 'saturat ed', 'minim ally', 't opologically', 'e xhibited'). Reference [7] is missing its page number.","section":"Whole manuscript, typography","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision","confidential_remarks":"This is a strong and clearly significant manuscript whose central comparison is sound at leading order. The requested changes are confined to stating the joint asymptotic regime explicitly and to typography; none of them requires new computation. I would be glad to see a revised version, and I do not expect to need further refereeing beyond checking that the regime statement has been added."},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"Short version: this is the paper that makes A/4 a counting result rather than an analogy, and the calculation is short enough to check line by line. It holds up. Send it out. What is new is the assembly, not the parts. The five-dimensional extremal solution carrying both axion and Ramond-Ramond charge, its horizon area in (2.10), and the observation that the D-brane bound states carrying those same charges have a degeneracy governed by Cardy growth — that combination is the contribution. The two sides are computed independently: gravity from the effective action, the count from the elliptic genus of a symmetric product. Neither uses the other, and A/4 is not an input to the count. That non-circularity is the whole point, and it survives scrutiny. Where I would push: the stress-test note is right that the two expansions are controlled in different regimes. Section 1 says large Q_H and fixed Q_F; section 2 says (2.11) is reliable only when both charges are large. Cardy as applied wants n >> c, i.e. Q_H >> 3 Q_F^2. So the honest statement is an overlap Q_H >> Q_F^2 >> 1 that the paper never writes down. Footnote 8 gestures at large-c corrections but does not tie them to the gravity-side expansion. This is a precision-of-framing issue, not an error — the leading terms do agree — but the abstract's unqualified 'is derived' [abstract:s0] reads stronger than the body's leading-order match. One sentence in revision would fix it. I disagree with the reader on one point: the symmetric-product identification is not the load-bearing assumption. Footnote 7 says explicitly that only dim M = 4k matters, and c = 6k follows from hyperkählerity, so the exponent does not care which D-brane configuration you pick. The reader's 6.0 on soundness is about right, but for the regime reason, not the sigma-model reason. The paper is candid elsewhere: the elliptic genus bounds rather than equals the degeneracy, and the discussion of strong coupling at large N is honest about where the brane picture breaks down. Who should read it: anyone working on black hole microstates, D-brane counting, or the precursors of AdS/CFT. It is a foundational reference, not a specialist curiosity. Recommendation: send it to referees; it will be cited for decades.","headline":"The paper that turns A/4 into a state count rather than a thermodynamic analogy; the derivation is short, independent on both sides, and holds up, with only a framing soft spot about the joint asymptotic regime.","tokens_in":10423,"tokens_out":1404,"would_cite":true,"duration_ms":21857,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":["04.70.Dy","11.25.-w","11.25.Mj"],"model":"selfhosted/deepseek-v4.1-flash","headline":"The horizon area of a five-dimensional extremal black hole equals a count of string bound states, factor for factor.","keywords":["black hole entropy","Bekenstein-Hawking area law","BPS states","D-branes","string theory compactification","symmetric product sigma model","elliptic genus","Cardy formula"],"falsifier":"Compute the degeneracy for a charge configuration outside the checked classes — for instance non-primitive cycles or a generic $Q_F$ — and find that its growth is not $2π√(Q_H(Q_F^2/2+1))$, or find that the leading large-charge correction to the horizon entropy from higher-derivative terms fails to match the corresponding correction to the state count.","tokens_in":10640,"feed_emoji":"🕳️","tokens_out":3107,"duration_ms":45154,"temperature":0.7,"pith_summary":"The paper sets out to close the loop on a twenty-year analogy: the laws of black hole mechanics look like thermodynamics, but nobody had counted the underlying states and recovered the one-quarter-area law from first principles. Strominger and Vafa take a class of extremal black holes in five noncompact dimensions that carry both an axion charge $Q_H$ and an electric charge $Q_F$, compute the horizon area from the low-energy effective action, and get $S_{BH} = 2π√(Q_H Q_F^2/2)$. They then count the BPS bound states carrying the same two charges using D-branes wrapped on cycles of $K3 × S^1$, and find the degeneracy grows as $S_{stat} = 2π√(Q_H(Q_F^2/2 + 1))$. For large charges the two expressions agree. If this holds, the area law for these holes is not an analogy but a statistical mechanical identity, and the black hole information question becomes a concrete computation in a D-brane worldvolume theory.","feed_headline":"Black hole entropy, derived by counting string states","feed_subtitle":"A five-dimensional extremal hole's horizon area matches a D-brane bound-state count, factor for factor.","key_machinery":"Two objects carry the argument. The first is the identification of the D-brane worldvolume theory on $S^1 × R$ with a supersymmetric $\\sigma$ model whose target is the symmetric product $(K3)^{[Q_F^2/2+1]}/S_{Q_F^2/2+1}$; this fixes the central charge $c = 6(Q_F^2/2 + 1)$ that sets the exponential growth. The second is the Cardy formula, which turns that central charge and the momentum $n = Q_H$ around the circle into an asymptotic degeneracy $d(n,c) ∼ exp(2π√(nc/6))$. The elliptic genus is the computable, moduli-independent stand-in for the degeneracy: it counts right-moving ground states with a sign, and the paper argues the sign weighting does not affect the leading growth. Momentum around $S^1$ becomes axion charge $Q_H$ after dualizing the circle, so the states being counted are exactly the ones the black hole solution describes.","core_discovery":"The central claim is that the Bekenstein-Hawking entropy of a five-dimensional extremal dyonic black hole is reproduced, to leading order in large charges, by counting BPS soliton bound states. On the gravity side, the near-horizon geometry is a charged Robinson-Berttoti universe with constant dilaton fixed by the charges; the horizon is a three-sphere of area $8π√(Q_H Q_F^2/2)$, giving $S_{BH} = 2π√(Q_H Q_F^2/2)$ (Eq. 2.11). On the microscopic side, the same charges are realized by D1-, D3- and D5-branes wrapping $S^1$ times holomorphic cycles of $K3$, and the relevant worldvolume theory on $S^1 × R$ is a supersymmetric $\\sigma$ model on the symmetric product of $Q_F^2/2 + 1$ copies of $K3$. Counting right-moving ground states with arbitrary left-moving excitation — the elliptic genus — gives a Cardy growth with central charge $c = 6(Q_F^2/2 + 1)$ at level $n = Q_H$, hence $S_{stat} = 2π√(Q_H(Q_F^2/2 + 1))$ (Eq. 3.4). The two formulas coincide to leading order for large $Q_F$ and $Q_H$, which is exactly the regime where the gravity computation is reliable.","pith_inferences":["The '+1' under the square root is doing visible work: at $Q_F = 0$ the horizon area vanishes and $S_{BH} = 0$, but the microscopic formula gives $2π√(Q_H)$, a nonzero count. That mismatch suggests the counting formula knows about states that the degenerate horizon does not, and that the correspondence is between two different limits of one object rather than an identity term by term.","Because the elliptic genus is a signed sum, the paper's count and a genuine state count can diverge at subleading order in $Q_H$. Computing that splitting would be a sharp test of how literally the symmetric-product picture should be taken.","If the microstates really are unordered points on $K3$, then infalling matter is delocalized over the whole horizon almost immediately, which is the microscopic shape a smooth, unitary answer to the information puzzle would have to take.","The same Cardy-counting template should apply to near-extremal holes obtained by adding a small amount of momentum or energy above extremality, where the growth is still governed by the same central charge but the horizon area is no longer exactly matched — a place to look for controlled corrections to the area law."],"forward_implications":["For this class of extremal holes the area law is a derived statistical identity rather than a thermodynamic analogy, so black hole dynamics and thermodynamics are the same laws, not merely similar ones.","The entropy formula is independent of the asymptotic value of the dilaton, because the near-horizon geometry is pinned by the charges; the result survives adiabatic changes of the asymptotic moduli.","The same derivation is claimed to go through for $N=8$ toroidal compactifications, with $T^4$ replacing $K3$ in the symmetric product, since the dimension of the resulting moduli space is unchanged.","The agreement requires both $Q_H$ and $Q_F$ to be large, since string-loop corrections are suppressed by $Q_F/Q_H$ and sigma-model corrections by inverse powers of the string-frame Schwarzschild radius.","The D-brane description supplies a concrete setting in which low-energy scattering off an extremal black hole could be computed, which is where the unitarity question for the information puzzle would be settled."],"supporting_citations":[{"why":"Supplies the conjecture that the D-brane worldvolume theory on $S^1 × R$ is a sigma model on the symmetric product of $Q_F^2/2+1$ copies of $K3$, the identification the whole count rests on.","marker":"[29]"},{"why":"Verifies that symmetric-product identification for restricted charge configurations, giving the paper its main evidence that the target space is right.","marker":"[30]"},{"why":"Extends the check to further brane configurations and generalizes the correspondence between spacetime BPS states and worldvolume BPS states used to set up the count.","marker":"[31]"},{"why":"Provides the Cardy formula for the asymptotic growth of left-moving oscillator degeneracies at large level, which converts the central charge into Eq. (3.4).","marker":"[39]"},{"why":"Establishes that D-branes are the sources of Ramond-Ramond charge, which is what makes the wrapped branes the right objects to carry $Q_F$.","marker":"[26]"},{"why":"Argues that spacetime BPS states correspond to supersymmetric ground states of the D-brane worldvolume theory, the principle licensing the state count.","marker":"[27]"},{"why":"Supplies the extremal five-dimensional charged black hole solution whose horizon area is computed on the gravity side.","marker":"[19]"},{"why":"Gives the four-dimensional dyonic solution obtained by reducing the five-dimensional one, and shows it is annihilated by a quarter of the supersymmetries.","marker":"[20]"}],"fun_headline_variants":["String states reproduce a black hole's area-entropy law, to leading order","Counting D-brane states matches a 5D extremal hole's entropy at large charge","Bekenstein-Hawking entropy matched by a D-brane count, but only at large charges","Elliptic genus on K3 grows as a horizon area law, factor for factor at leading order","The match is not exact: string counting equals horizon entropy only in the large-charge li"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The count assumes that the D-brane worldvolume theory for generic charges really is the supersymmetric $\\sigma$ model on the symmetric product of $Q_F^2/2+1$ copies of $K3$; that identification was conjectured and only checked for restricted charge configurations, and the central charge driving the exponential growth comes entirely from it.","fun_headline_variants_meta":{"raw":{"variants":["String states reproduce a black hole's area-entropy law, to leading order","Counting D-brane states matches a 5D extremal hole's entropy at large charge","Bekenstein-Hawking entropy matched by a D-brane count, but only at large charges","Elliptic genus on K3 grows as a horizon area law, factor for factor at leading order","The match is not exact: string counting equals horizon entropy only in the large-charge limit"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":1218,"prompt_tokens":860,"completion_tokens":358,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":255}},"tokens_in":860,"tokens_out":358,"duration_ms":4843,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":255,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-10-29T19:21:48.111606+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"Compute the degeneracy for a charge configuration outside the checked classes — for instance non-primitive cycles or a generic $Q_F$ — and find that its growth is not $2π√(Q_H(Q_F^2/2+1))$, or find that the leading large-charge correction to the horizon entropy from higher-derivative terms fails to match the corresponding correction to the state count.","supporting_citations":[{"cited_title":"Gas of D-Branes and Hagedorn Density of BPS States","cited_arxiv_id":"hep-th/9511088","evidence_quote":"Supplies the conjecture that the D-brane worldvolume theory on $S^1 × R$ is a sigma model on the symmetric product of $Q_F^2/2+1$ copies of $K3$, the identification the whole count rests on."},{"cited_title":"D-Branes and Topological Field Theories","cited_arxiv_id":"hep-th/9511222","evidence_quote":"Verifies that symmetric-product identification for restricted charge configurations, giving the paper its main evidence that the target space is right."},{"cited_title":"Instantons on D-branes","cited_arxiv_id":"hep-th/9512078","evidence_quote":"Extends the check to further brane configurations and generalizes the correspondence between spacetime BPS states and worldvolume BPS states used to set up the count."},{"cited_title":"Cardy, Nucl","cited_arxiv_id":null,"evidence_quote":"Provides the Cardy formula for the asymptotic growth of left-moving oscillator degeneracies at large level, which converts the central charge into Eq. (3.4)."},{"cited_title":"Dirichlet-Branes and Ramond-Ramond Charges","cited_arxiv_id":"hep-th/9510017","evidence_quote":"Establishes that D-branes are the sources of Ramond-Ramond charge, which is what makes the wrapped branes the right objects to carry $Q_F$."},{"cited_title":"Bound states of strings and p-branes","cited_arxiv_id":"hep-th/9510135","evidence_quote":"Argues that spacetime BPS states correspond to supersymmetric ground states of the D-brane worldvolume theory, the principle licensing the state count."},{"cited_title":"Gibbons, Nucl","cited_arxiv_id":null,"evidence_quote":"Supplies the extremal five-dimensional charged black hole solution whose horizon area is computed on the gravity side."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the four-dimensional dyonic solution obtained by reducing the five-dimensional one, and shows it is annihilated by a quarter of the supersymmetries."}],"review_version":1}