arXiv subjects

Tong Arthur Wu

Publications and source records attributed to Tong Arthur Wu.

10 recordsLinked to original sources

arXiv2026

MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks

Existing evaluations of agents with memory typically assess memorization and action in isolation. One class of benchmarks evaluates memorization by testing recall of past conversations or text but fails to capture how memory is used to guide future decisions. Another class focuses on agents acting in single-session tasks without the need for long-term memory. However, in realistic settings, memorization and action are tightly coupled: agents acquire memory while interacting with the environment, and subsequently rely on that memory to solve future tasks. To capture this setting, we introduce MemoryArena, a unified evaluation gym for benchmarking agent memory in multi-session Memory-Agent-Environment loops. The benchmark consists of human-crafted agentic tasks with explicitly interdependent subtasks, where agents must learn from earlier actions and feedback by distilling experiences into memory, and subsequently use that memory to guide later actions to solve the overall task. MemoryArena supports evaluation across web navigation, preference-constrained planning, progressive information search, and sequential formal reasoning, and reveals that agents with near-saturated performance on existing long-context memory benchmarks like LoCoMo perform poorly in our agentic setting, exposing a gap in current evaluations for agents with memory. MemoryArena is now released at https://memoryarena.github.io/.

cs.CL↗

arXiv2026

Reviving WIMP dark matter with temperature-dependent couplings

The persistent null results at dark matter (DM) direct-detection experiments have pushed the popular weakly interacting massive particle (WIMP) DM to tight corners. Generic WIMP models with direct-detection rate below the current upper limits often lead to a thermally overproduced relic abundance after freeze-out. To resolve this conundrum, we propose a novel scenario where DM has temperature-dependent couplings with the standard model (SM) bath. A scalar field having a large vacuum expectation value (VEV) at high temperatures generates sizeable DM-SM interactions leading to efficient DM annihilations responsible for generating the desired thermal relic. At lower temperatures, the scalar field VEV settles down to a small value as a result of a phase transition which can generically be of first order, effectively leading to suppressed DM-SM interaction rate at low temperature, consistent with null results at direct-detection experiments. Upper bound on thermal DM mass forces the first-order phase transition (FOPT) to occur at scales such that the corresponding gravitational wave signal remains within reach of future experiments like LISA.

hep-ph↗

arXiv2025

LHC Shines on Positivity

We show that hadron colliders have an excellent reach for positivity tests on a class of diphoton operators. Due to the helicity selection rules, the relevant dimension-6 operators either do not contribute or are highly constrained by other experimental observables. We show, for the first time, that the LHC can probe the positivity of the dimension-8 operators involving colored particles. The kinematic differential distributions of the diphoton final states are exploited to perform the $χ^2$ analysis. Through a global fit, the effective scale for these operators can be inclusively probed up to around 2 TeV at HL-LHC and over 5 TeV at future 100 TeV FCC-hh at 95% C.L., providing a powerful test of the positivity bounds up to multi-TeV scale.

hep-ph↗

arXiv2025

Muon collider probes of the gluonic quartic gauge couplings

We investigate the dimension-8 gluonic quartic gauge couplings (gQGCs) at future high-energy muon colliders through the process $μ^{+}μ^{-}\!\to ggγ$. Using detailed event simulation and optimized kinematic selections, we derive projected sensitivities to the Wilson coefficients and their associated new-physics scales, showing that muon colliders can probe deep into the multi-TeV regime and significantly surpass current LHC bounds. We further present the positivity bounds on those Wilson coefficients, as theoretical constraints from the fundamental principles of quantum field theory. Our results establish $μ^{+}μ^{-}\!\to ggγ$ as one of the most sensitive probes of dimension-8 new physics at future muon colliders.

hep-ph↗

arXiv2025

Quantum Information meets High-Energy Physics: Input to the update of the European Strategy for Particle Physics

Some of the most astonishing and prominent properties of Quantum Mechanics, such as entanglement and Bell nonlocality, have only been studied extensively in dedicated low-energy laboratory setups. The feasibility of these studies in the high-energy regime explored by particle colliders was only recently shown and has gathered the attention of the scientific community. For the range of particles and fundamental interactions involved, particle colliders provide a novel environment where quantum information theory can be probed, with energies exceeding by about 12 orders of magnitude those employed in dedicated laboratory setups. Furthermore, collider detectors have inherent advantages in performing certain quantum information measurements, and allow for the reconstruction of the state of the system under consideration via quantum state tomography. Here, we elaborate on the potential, challenges, and goals of this innovative and rapidly evolving line of research and discuss its expected impact on both quantum information theory and high-energy physics.

hep-ph↗

arXiv2025

Entanglement and Bell Nonlocality in $τ^+ τ^-$ at the LHC using Machine Learning for Neutrino Reconstruction

Experiments at the CERN Large Hadron Collider (LHC) have accumulated an unprecedented amount of data corresponding to a large variety of quantum states. Although searching for new particles beyond the Standard Model of particle physics remains a high priority for the LHC program, precision measurements of the physical processes predicted in the Standard Model continue to lead us to a deeper understanding of nature at high energies. We carry out detailed simulations for the process $pp \to τ^+τ^- X$ to perform quantum tomography and to measure the quantum entanglement and the Bell nonlocality of the $τ^+τ^-$ two qubit state, including both statistical and systematic uncertainties. By using advanced machine learning techniques for neutrino momentum reconstruction, we achieve precise measurements of the full spin density matrix, a critical advantage over previous studies limited by reconstruction challenges for missing momenta. Our analysis reveals a clear observation of Bell nonlocality with high statistical significance, surpassing 5$σ$, establishing $τ^+ τ^-$ as an ideal system for quantum information studies in high-energy collisions. Given its experimental feasibility and the high expected sensitivity for Bell nonlocality, we propose that $τ^+ τ^-$ should be regarded as the new benchmark system for quantum information studies at the LHC, complementing and extending the insights gained from the $t\bar{t}$ system.

hep-ph↗

arXiv2025

Quantum Tomography in Neutral Meson and Antimeson Systems

The flavor space of particles produced in collider environments contains informative quantum correlations. We present a systematic approach for constructing the complete flavor density matrix for a meson and antimeson system ($M \bar M$) in the Bloch vector space at a given time $t$, which can be at or after production. We point out that the $B_s^0$ and $K^0$ systems are superior to the $B^0_d$ and $D^0$ systems for quantum tomography because of their flavor oscillation and decay properties. Performing quantum tomography for the $M \bar M$ system can facilitate the study of production mechanisms, decoherence phenomena, quantum information variables, and potential new sources of CP violation.

hep-ph↗

arXiv2025

Colorful Particle Production at High-Energy Muon Colliders

A high-energy $μ^+μ^-$ collider provides a wide variety of mechanisms for the production of new heavy particles. While the reach for such particles via the direct annihilation of $μ^+μ^-$ will approach the center-of-mass energy of the collider, the partonic fusions from gauge bosons, quarks, and gluons, originating from the incoming muon beams will open new channels for single production and pair production of particles with different quantum numbers. We present the production rates for a wide variety of colored states including color-triplets, color-sextets, color-octets, leptoquarks, and leptogluons. We find that pair production from the direct annihilation of $μ^+μ^-$ generally has a cross section of $0.1 - 1$ fb once above the production threshold. On the other hand, pair production through the quark and gluon content of the muon leads to a cross section of roughly $10^{-4}$ fb at the same particle mass. We perform simple estimations of the mass reach for each particle and find that a 10 TeV muon collider can extend the reach for color-triplets beyond what is possible at the high luminosity run of the Large Hadron Collider. Leptoquarks and leptogluons, with sensitivity driven by single production, can also be probed to higher masses at a muon collider than what the Large Hadron Collider can reach. A final example where a muon collider has superior reach is for color-octet scalars and vectors. Together, these cases illustrate the point that a muon collider is a competitive machine for searching for colored heavy particles, thus strengthening the motivation for such lepton colliders in the energy frontier. Although our study is focused on a muon collider, our results are largely applicable to high-energy $e^+e^-$ collisions as well.

hep-ph↗

arXiv2024

Capping the positivity cone: dimension-8 Higgs operators in the SMEFT

SMEFT Wilson coefficients are subject to various positivity bounds in order to be consistent with the fundamental principles of S-matrix. Previous bounds on dimension-8 SMEFT operators have been obtained using the positivity part of UV partial wave unitarity and form a (projective) convex cone. We derive a set of linear UV unitarity conditions that go beyond positivity and are easy to implement in an optimization scheme with dispersion relations in a multi-field EFT. Using Higgs scattering as an example, we demonstrate how to obtain closed bounds in the space of the three relevant dimension-8 coefficients, making use of the UV unitarity conditions as well as so-called null constraints that arise from full crossing symmetry. Specifically, we show that they are bounded by inequalities schematically going like $C<O\left((4π)^2\right)$. We compare the newly obtained upper bounds with the traditional perturbative unitarity bounds from within the EFT, and discuss some phenomenological implications of the two-sided positivity bounds in the context of experimental probes of Vector Boson Scattering.

hep-ph↗

arXiv2023

Quantum Entanglement and Bell Inequality Violation in Semi-Leptonic Top Decays

Quantum entanglement is a fundamental property of quantum mechanics. Recently, studies have explored entanglement in the $t\bar{t}$ system at the Large Hadron Collider (LHC) when both the top quark and anti-top quark decay leptonically. Entanglement is detected via correlations between the polarizations of the top and anti-top and these polarizations are measured through the angles of the decay products of the top and anti-top. In this work, we propose searching for evidence of quantum entanglement in the semi-leptonic decay channel where the final state includes one lepton, one neutrino, two $b$-flavor tagged jets, and two light jets from the $W$ decay. We find that this channel is both easier to reconstruct and has a larger effective quantity of data than the fully leptonic channel. As a result, the semi-leptonic channel is $60\%$ more sensitive to quantum entanglement and a factor of 3 more sensitive to Bell inequality violation, compared to the leptonic channel. In $139~{\rm fb}^{-1}$ ($3~{\rm ab}^{-1}$) of data at the LHC (HL-LHC), it should be feasible to measure entanglement at a precision of $\lesssim 3\%\ (0.7\%)$. Detecting Bell inequality violation, on the other hand, is more challenging. With $300~{\rm fb}^{-1}$ ($3~{\rm ab}^{-1}$) of integrated luminosity at the LHC Run-3 (HL-LHC), we expect a sensitivity of $1.3σ$ ($4.1 σ$). In our study, we utilize a realistic parametric fitting procedure to optimally recover the true angular distributions from detector effects. Compared to unfolding this procedure yields more stable results.

hep-ph↗