{"total":19,"items":[{"citing_arxiv_id":"2606.22215","ref_index":152,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Accelerating Discovery: Model-Agnostic Likelihoods for the Reinterpretation of Particle Physics Results and their Application to the Belle II $B^{+}\\to K^{+}\\nu\\bar{\\nu}$ Measurement","primary_cat":"hep-ex","submitted_at":"2026-06-20T20:31:23+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A reweighting method creates model-agnostic likelihoods from histogram analyses, applied to Belle II B+ to K+ nu nubar data for WET constraints and light new physics searches.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.20036","ref_index":94,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Evolving Dark Energy Is Vacuum Energy After All","primary_cat":"astro-ph.CO","submitted_at":"2026-06-18T10:10:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A QCD-vacuum-based model of dynamical dark energy fits Planck+ACT+SPT, DESI DR2, and supernova data while reproducing the late-time evolution favored by DESI.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.02740","ref_index":13,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"ScoreStop: Gradient-based early stopping using functional score tests","primary_cat":"stat.ML","submitted_at":"2026-06-01T18:05:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"ScoreStop introduces a functional score test for early stopping in gradient boosting, testing the null that the current predictor minimizes population risk with a scale-invariant statistic of known asymptotic distribution.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.02699","ref_index":231,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A New Record Census of Dwarf AGN and a Bimodal $M_{\\rm BH}$-$M_{\\star}$ Scaling Relation with DESI DR1","primary_cat":"astro-ph.GA","submitted_at":"2026-06-01T18:00:00+00:00","verdict":"CONDITIONAL","verdict_confidence":"UNKNOWN","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DESI DR1 yields 314k high-mass and 9.6k dwarf AGN, extending the M_BH-M_star relation to log M_star ~7.8 and suggesting two evolutionary pathways for galaxies and black holes.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.18958","ref_index":50,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Directly tracking the re-brightening of a supermassive black hole accretion disk","primary_cat":"astro-ph.HE","submitted_at":"2026-05-18T18:00:05+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Six years of UV/X-ray monitoring of ESO 511-G030 reveals a >10x UV rise from the accretion disk, tight disk-corona coupling above ~1% Eddington, and breakdown below, indicating an accretion-state transition.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.14245","ref_index":88,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Constraints on anomalous Higgs boson couplings to vector bosons and fermions using the $\\gamma\\gamma$ final state in proton-proton collisions at $\\sqrt{s}$ = 13 TeV","primary_cat":"hep-ex","submitted_at":"2026-05-14T01:22:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Updated constraints from the full Run 2 dataset show no evidence for anomalous Higgs couplings in gluon fusion, vector boson fusion, and associated production modes decaying to two photons.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"In the Hgg analysis the POI isf ggH a3 , whilef a3 is fixed to the SM value (fa3 =0). The parameter ⃗ˆfrepresents the best fit values of the POIs that maximize the likelihood. In all fits,m H is fixed to its most precisely measured value of 125.38 GeV [84]. In this analysis, as in many other effective field theory studies, the conditions of Wilks' theo- rem [88] are violated because of the quadratic dependence of the signal yield on the coupling parameters [89, 90], which prevents a correct estimation of the confidence level (CL) using the likelihood-ratio test statistic. The likelihood ratio intervals for the quadratic model may thus undercover or overcover. Nevertheless, for consistency with previous analyses, we quote con-"},{"citing_arxiv_id":"2605.11063","ref_index":29,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Electronic Direct Detection of Light Dark Matter with Intermediate-Mass Mediators","primary_cat":"hep-ph","submitted_at":"2026-05-11T18:00:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"For sub-GeV dark matter, the light and heavy mediator mass limits in direct detection are separated by up to three orders of magnitude in mediator mass, enabling precise sensitivity calculations for Si, Ge, and DAMIC-M targets.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"iments typically report their energy thresholds in terms of the number of electron-hole pairs, or ionization charge, Q, produced in a scattering event. For an energy depo- sitionω, this is Q= 1 + \u0016 ω−E g ϵ \u0017 ,(13) whereϵis the energy required for each addi- tional electron-hole pair created. We assumeϵ= 3.6 eV (2.9 eV) andE g = 1.11 eV (0.67 eV) for a Si (Ge) target [29, 32-34]. An energy threshold equal to the band gap corresponds to including all electronic transi- tions withQ≥1; largerω th increases the lower bound on Qaccording to Eq. (13), and reduces the sensitivity for all DM and mediator masses. Therefore increasingω th will increase the ratio ¯σe/¯σFI e in Figs. 2 and 3, thereby decreasing the experimental sensitivity."},{"citing_arxiv_id":"2605.08471","ref_index":39,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Asymptotics for likelihood ratio tests of boundary points with singular information and unidentifiable nuisance parameters","primary_cat":"math.ST","submitted_at":"2026-05-08T20:44:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":8.0,"formal_verification":"none","one_line_summary":"The LRT statistic converges in distribution to the supremum of a bar-chi-squared process under the null and a noncentral version under local alternatives, with the same form whether or not the information matrix is singular due to the nuisance parameter.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"depends on t and t′ through (τ, K) and (τ ′, K′), the same is true for ρ(t, t′), and therefore the limit process X(t) will not depend on p. Hence we regard X(τ, K) as a function of (τ, K) only, and the asymptotic LRT statistic reduces to Λ = sup τ ∈[0,T ], K∈[ϵ,1−ϵ] X(τ, K). There are r = 2 constraints that define ∆(t). From this and ( 9) it follows that X(τ, K) \u0018 0.5 w2(τ, K) \u0001 χ2 0 + 0.5 χ2 1 + w2(τ, K)χ2 2, (39) has a ¯χ2-distribution under H0 for each (τ, K), with w2(τ, K) = cos −1(I12(τ, K)/ p I11(τ, K)I22(τ, K))/(2π). (40) The weight ( 40) of the χ2 2 component in ( 39) can be deduced from ( 25). Indeed, since C = [0 , 1)2, we may take V (t) = Id2 in ( 24), giving ui(t) = [ A(t)−T]i (the i-th row of A(t)−T). Then ui(t)uj(t)T = [I(t)−1]ij = I −1 ij (τ, K), and consequently"},{"citing_arxiv_id":"2605.08034","ref_index":13,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Semiparametric Efficient Test for Interpretable Distributional Treatment Effects","primary_cat":"stat.ML","submitted_at":"2026-05-08T17:23:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"DR-ME is the first semiparametrically efficient finite-location kernel test for interpretable distributional treatment effects, using orthogonal doubly robust features derived from observational data.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"For fixed locations, we study quadratic-mean-differentiable local submodels through the finite-location null, following the local asymptotic testing perspective of Neyman and Pearson[25], Le Cam and Yang[20], and van der Vaart[36]. The canonical-gradient statistic attains the efficient finite-signal Gaussian experiment. Consequently, the Hotelling statistic [13] has a chi-square null limit and a noncentral chi-square local limit, with power governed by the interventional drift whitened by the canonical-gradient covariance. This whitening is not an accidental consequence of plugging a doubly robust score into a conventional Hotelling statistic; it is the covariance geometry of the efficient finite-signal experiment."},{"citing_arxiv_id":"2605.00398","ref_index":48,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"M-CaStLe: Uncovering Local Causal Structures in Multivariate Space-Time Gridded Data","primary_cat":"cs.LG","submitted_at":"2026-05-01T04:40:35+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"M-CaStLe generalizes local stencil-based causal discovery to the multivariate case and decomposes resulting graphs into reaction and spatial components for interpretation in space-time gridded data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.21857","ref_index":63,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Odd Physics Off the Diagonal: Constraining CP-violating SMEFT with Quantum Tomography","primary_cat":"hep-ph","submitted_at":"2026-04-23T16:51:52+00:00","verdict":null,"verdict_confidence":null,"novelty_score":null,"formal_verification":null,"one_line_summary":null,"context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.20632","ref_index":63,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Review: A new method for estimation and use of systematic errors in Poisson regression","primary_cat":"astro-ph.IM","submitted_at":"2026-04-22T14:45:17+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"The reviewed method generalizes the Cash statistic C_min and likelihood-ratio ΔC to include systematic uncertainties in Poisson data, allowing simultaneous assessment of systematics level and model goodness-of-fit.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.17926","ref_index":32,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Information on hidden birth events restores identifiability in phylodynamic inference","primary_cat":"q-bio.PE","submitted_at":"2026-04-20T08:06:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":8.0,"formal_verification":"none","one_line_summary":"Hidden birth event information restores identifiability to time-dependent birth-death phylodynamic models; mutation-at-birth models make sequences sufficient to recover it.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.15598","ref_index":70,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"When do trajectories matter? Identifiability analysis for stochastic transport phenomena","primary_cat":"nlin.CG","submitted_at":"2026-04-17T00:37:47+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Trajectory data resolves structural non-identifiability in lattice random walk diffusion models that count data alone cannot, with analysis of experimental design impacts on practical identifiability.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"When parameters are well-identified by the data we obtain a loglikelihood func- tion with a unique MLE and a relatively constrained confidence set where¯ℓ≥ ¯ℓ∗. In contrast, poorly identified and non-identifiable problems are characterised by a poorly defined MLE that is potentially non-unique, and a relatively broad 95% parameter confidence set, and/or strong correlation between estimates of different parameters [70]. We will show examples of both cases as we proceed. For problems that involve three parameters, e.g.θ= (D, v, K) ⊤ when we consider biased motility, we are no longer able to simply visualize ¯ℓ c(θ|N(t)) as a two-dimensional heat map or contour plot. In these cases we construct a series of profile likelihood functions by partitioningθinto interest parametersψ, and nuisance parametersω, so thatθ="},{"citing_arxiv_id":"2604.09397","ref_index":54,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A TeV-based Determination of the Local Extragalactic Background Light and its Consistency with Galaxy Counts and Direct Measurements","primary_cat":"astro-ph.HE","submitted_at":"2026-04-10T15:11:13+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"TeV gamma-ray optical depths yield a local EBL intensity consistent with integrated galaxy light to within 25% over 0.5-30 microns and incompatible at 3-5 sigma with near-IR excesses from IRTS and CIBER.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"likelihood maximization fit forαvalues ranging from 0 to 2.5, with a step size of 0.05. By multiplying the indi- vidual likelihoods across all sources within a given bin, we constructed a global profile likelihood forα. The global profile likelihoodL(α) yieldsα best, whereLis maximized, representing the best-fit EBL density rela- tive to the model. According to Wilks' theorem (Wilks 1938), under the null hypothesis and asymptotic condi- tions, the test statisticT S=−2 log Λ, where Λ is the likelihood ratioL(α)/L(α best), follows aχ 2 distribution with one degree of freedom. This allows the determina- tion of the 1σconfidence interval, corresponding to the values ofαfor which ∆(−2 log Λ) = 1. The above procedure was performed for seven EBL"},{"citing_arxiv_id":"2603.18223","ref_index":106,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Gamma-Ray Bursts as an Independent High-Redshift Probe of Dark Energy","primary_cat":"astro-ph.CO","submitted_at":"2026-03-18T19:14:57+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Forecasts show that ~66 optical GRBs can achieve σ_w ≈ 0.47 in wCDM using Dainotti relations, matching Planck precision and enabling independent high-redshift tests of dark energy.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"(1329) X-ray (optical) sources. Nonetheless, for samples of 4000 GRBs in both X-ray and optical cases, we obtain contours closed outside theΛCDM pointw 0 =−1,w a = 0within3σ. While this is encouraging, it does not imply that thew0waCDM model is favoured overΛCDM at the 3σlevel. To compare two models in such a scenario, one would need to apply Wilks' theorem (S. S. Wilks 1938). However, this is not applicable here, because our posteriors indicate only an upper limit onwa. In such a case, to compare the two models, we introduce a pivot parameterwp, which decorrelates thew0 andw a parameters: wp =w 0 + (1−a p)wa,(15) wherea p is the pivot scale factor, chosen in a way which makeswp andw a uncorrelated and minimizes the uncertainty"},{"citing_arxiv_id":"2512.23458","ref_index":105,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Observations of the Fermi bubbles and the Galactic center excess with the DArk Matter Particle Explorer","primary_cat":"astro-ph.HE","submitted_at":"2025-12-29T13:34:23+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DAMPE independently detects the Fermi bubbles at 26 sigma and the Galactic center GeV excess at 7 sigma, with the excess spectrum and morphology matching Fermi-LAT and fitting a 50 GeV dark matter particle annihilating to b quarks.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"blue points, exhibiting a hard spectrum with a cutoff at ∼ 100 GeV. The test statistic (TS) of the source is TS ≡ − 2 ∑ i ln( ˆLnull,i/ ˆLalt,i) = 757 .4, where ˆLnull,i and ˆLalt,i are the likelihood values for the optimal null and alternative models in the i-th SED bin. Since the TS value follows the χ2 distribution with 11 degrees of freedom in the null hypothesis ( S. S. Wilks 1938), the significance of the bubbles is 26.5σ. We estimate the systematic uncertainty of the spec- trum induced by inaccurate background models. Firstly, the GDE model is tested. The Galprop GDE model is derived assuming that the CR particles are injected fol- lowing a CR source distribution, propagate throughout the Galaxy within a cylindrical propagation halo, and"},{"citing_arxiv_id":"2505.07698","ref_index":14,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Likelihood Ratio Framework for Highly Motivated Subdominant Signals","primary_cat":"hep-ph","submitted_at":"2025-05-12T16:05:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Develops a likelihood ratio framework for assessing highly motivated subdominant new physics signals against experimental residuals that appear consistent with background predictions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2309.00267","ref_index":24,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback","primary_cat":"cs.CL","submitted_at":"2023-09-01T05:53:33+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RLAIF matches RLHF on summarization and dialogue tasks, with a direct-RLAIF variant achieving superior results by using LLM rewards directly during training.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}