150813 research outputs found
Sort by
Tracing the stepwise Darwinian evolution of a plant halogenase.
No full textBiohalogenation is rare in plant metabolism, with the Menispermaceae's chloroalkaloid acutumine being an exception. This involves a specialized dechloroacutumine halogenase (DAH) from the iron- and 2-oxoglutarate-dependent dioxygenase (2ODD) family. While DAH is presumed to have evolved from an ancestral 2ODD, how enzyme specialization arises through Darwinian processes remains a fundamental question in understanding metabolic evolution. Here, we investigate the evolutionary history of DAH using the chromosomal-level genome of Menispermum canadense. Phylogenomic dating and synteny analyses reveal DAH evolution through tandem duplication of an ancestral flavonol synthase (FLS) gene, followed by neofunctionalization and gene loss events. Structural modeling, molecular dynamics, and site-directed mutagenesis identify mutations enabling the catalytic switch from FLS to DAH. This required traversing a complex evolutionary landscape with deep fitness valleys separating intermediate states captured in the M. canadense genome. Our findings illustrate how enzymatic functions evolve through lineage-specific pathways, reshaping active sites and enabling catalytic mechanism-switching mutations
How FDI reshapes host markets’ trade profile and politics
No full textA fast-growing literature indicates that firms’ engagement in foreign directinvestment (FDI) and trade is key to understanding deepening global valuechains and their political implications. However, existing studies have mainlyfocused on the ramifications for FDI home countries while often overlookingthe firm-product level interactions between FDI and trade, where their inter-dependencies manifest. This study examines how firms’ FDI reshapes hostcountries’ trade profiles at this level, empowering new political coalitions fortrade liberalization. Analyzing greenfield FDI projects globally since 2003, wefind that hosts experienced an average increase of over 45 export products inthe following year. To overcome the challenges of connecting firms to prod-ucts, we link FDI data with Vietnamese customs records. We find that Viet-namese export (import) volumes of FDI-related products increased by 90%(30%) within 4 years of initial investments. Importantly, these products alsobenefited from more substantial tariff cuts in bilateral Free Trade Agreements
Online Acquisition of Simulatable Rigid Object Models
No full textHow can we build a robot that operates autonomously in a home environment over long periods of time? A key requirement is the ability to perceive and understand its surroundings, including the objects it will interact with. This thesis investigates how a robot can reconstruct previously unknown objects and integrate them into a physics simulation for planning. We explore two methods for reconstructing the 3D geometry of objects and test their performance in simulation and in real-world experiments. Our results demonstrate that a learned depth model enables 3D reconstruction of unknown objects and their successful integration into simulation environments. Additionally, we investigate methods for estimating an object’s inertial parameters, using its reconstructed mesh and through manipulation.M.Eng
A unified semantics for distributive and non-distributive universal quantifiers across languages
No full textUniversal quantifiers differ in whether they are restricted to distributive interpretations, like English every, or permit non-distributive interpretations, like English all. This interpretational difference is traditionally captured by positing two unrelated lexical entries for distributive and non-distributive quantification. But this lexical approach does not explain why distributivity correlates with number: cross-linguistically, distributive universal quantifiers typically take singular complements, while non-distributive quantifiers consistently take plural complements. We derive this correlation by proposing a single lexical meaning for the universal quantifier, which derives a non-distributive interpretation if the restrictor predicate is closed under sum, but a distributive interpretation if it is quantized. Support comes from languages in which the same lexical item expresses distributive or non-distributive quantification depending on the number of the complement. For languages like English that have different expressions for non-distributive and distributive quantification, we propose that the distributive forms contain an additional morphosyntactic element that is semantically restricted to combine with a predicate of atomic individuals. This is motivated by the fact that in several languages, the distributive form is structurally more complex than the non-distributive form and sometimes even contains it transparently. We further show that in such languages, there are empirical advantages to taking the choice between distributive and non-distributive quantifier forms to be driven by semantic properties of the restrictor predicate, rather than morphosyntactic number
Initial checkout of the Psyche electric propulsion system
No full textNASA’s Psyche spacecraft launched on October 13, 2023, and soon afterward the mission operations team began spacecraft initial checkout activities. For the electric propulsion system, the feed system and thruster gimbals were first prepared and then the rest of the subsystem completed an initial operations test during thruster bakeout. Thrust for each thruster was measured across the full range of operating powers and was in good agreement with pre-flight expectations. A weeklong test of the spacecraft and mission operations plan during thrusting activities was successful, but a thruster burn-in phenomenon was observed during full power operation that was longer than expected based on previous flight history. Data accumulated during the initial checkout activities shows that this burn-in behavior is different for each thruster and suggests that it is a result of the thruster discharge transitioning between two different plasma modes that can be mitigated by reducing discharge power and by adjusting the thruster magnet current. At the conclusion of the checkout activities, the subsystem had accumulated 357 h of thrusting operations while consuming 18.5 kg of propellant and was fully ready to begin the cruise phase of the mission
National crop field delineation for the United States
No full textComprehensive and accurate crop field boundary maps are crucial for digital agriculture, land management, and environmental monitoring. However, no high-quality field boundary dataset is publicly available in the United States. This thesis addresses this gap by creating a new, large dataset and training a deep learning model capable of mapping field boundaries. We built a dataset of over 15,000 image-mask pairs using high-resolution National Agriculture Imagery Program (NAIP) satellite imagery and curated field boundary labels. This dataset covers a variety of leading agricultural states and includes images taken at different scales to capture a wide variety of field sizes and layouts. We used this dataset to train an adapted ResUNet++ neural network model designed to segment crop fields. The trained model achieved around 0.8 for pixel-level accuracy, showing it can generally identify field areas well. However, its performance in matching predicted individual field instances with the ground truth instances (measured by mean instance Intersection over Union, or mIoU) was around 0.5. This lower instance score was largely due to the post-processing step, which converts the model’s probability predictions into separate field instances. Despite this, the field polygons produced by our approach are visually coherent with satellite field images and can be readily used with geospatial tools like Google Earth Engine. Our work provides a practical starting point for future research on mapping fields across the contiguous U.S. Potential directions for improvements may involve developing sharper boundary predictions, exploring direct instance segmentation models, refining post-processing methods, and expanding the dataset to include more challenging areas.M.Eng
Talk to the Hand: an LLM-powered Chatbot with Visual Pointer as Proactive Companion for On-Screen Tasks
No full textCHI ’25, Yokohama, JapanThis paper presents Pointer Assistant, a novel human-AI interaction technique for on-screen tasks. The design features a chatbot displayed as an extra mouse pointer, alongside the user’s, which proactively gives feedback on user actions while directing them to relevant areas on the screen and responding to the user’s direct chat messages. The effectiveness of the design’s key characteristics, pointer form and proactivity, was investigated in a study involving 220 participants in a financial budget planning task. Results demonstrated that the pointer design and interaction reduced task load while improving satisfaction with the experience, and increased the number of budget categories ideated during the task compared to the traditional passive chat log design. Participants viewed Pointer Assistant as a fun, innovative, and helpful visual guide while noting that its assertiveness can be improved. Future developments could offer even further enhancements to the user experience of human-AI collaboration and task outcomes
Succinct Cryptography via Propositional Proofs
No full textThe goal in modern cryptography is to obtain security while minimizing the use of computational resources. In recent years, we have been incredibly successful in our pursuit for efficiency, even for cryptographic tasks that were thought to be “science fiction”. For example, we have constructions of fully homomorphic encryption and private information retrieval from standard, cryptographic assumptions which achieve the ideal levels of succinctness. However, there are still some tasks in cryptography where achieving the “ideal” efficiency from standard assumptions has evaded us. In this thesis, we study the problem of achieving succinctness in two such settings: • Can we construct succinct indistinguishability obfuscation (IO) for Turing machines? In particular, can we construct an obfuscated program whose size is independent of the input length? • Can we construct succinct non-interactive arguments (SNARGs) for all of NP? While the problems seem unrelated at first glance, the root difficulty seems to stem from a similar place: both primitives have non-falsifiable security definitions. In fact, this type of barrier exists for many other cryptographic primitives, including witness encryption. This leads to a central question which we refer to as the “non-falsifiability barrier”: how can we construct non-falsifiable primitives from falsifiable assumptions? In this thesis, we show how to leverage propositional proofs to overcome the non-falsifiability barrier, and make substantial progress in the goal of achieving succinctness in both settings. Our main result is universal construction of both SNARGs and succinct IO for Turing machines from standard assumptions using propositional proofs. We then show several applications, including rate-1 IO for many programs, the first succinct secret sharin schemes for monotone circuits, and many more. Our results establish propositional proofs as a foundational tool for achieving succinctness across a broad range of cryptographic settings.Ph.D
A Close Look at RMP Entry Caching and Its Security Implications in SEV-SNP
No full textAMD’s Secure Encrypted Virtualization (SEV) technology is a pivotal component in AMD server processors that boosts cloud computing security. It achieves this by offering transparent memory encryption and managing keys for protecting virtual machines (VMs),
independently of the hypervisor’s trustworthiness. The latest iteration, SEV-Secure Nested Paging (SEV-SNP), introduces memory
integrity protection through a data structure called the Reverse
Map Table (RMP), which maps system physical addresses to guest
physical addresses and tracks ownership of physical pages.
The RMP is maintained in a dedicated region in DRAM. As every memory write triggers a check against an RMP entry, caching
RMP entries is crucial to alleviating the RMP’s performance impact. However, caching may create new security challenges, as it
can introduce new microarchitectural side-channels. In addition,
maintaining cache coherence is crucial for the RMP’s security guarantees. However, so far, neither the details of the RMP’s caching
behavior nor its security implications have been explored. This
paper aims to fill this gap by conducting a systematic study of the
RMP’s caching behavior. Through reverse engineering, we identify
that the RMP is not only cached in the TLB, but also in the L1D
and L2 data cache. Interestingly, this caching depends on the access
type on Zen 5. We also uncover the mechanisms by which cache
coherence across the TLB is enforced. We find that each update to
the RMP table triggers a global TLB flush across all cores. Finally,
we present several potential security implications and demonstrate
that an attacker can exploit RMP’s caching to leak physical address
information. A user process can leak 6 bits of the Physical Frame
Number (PFN) of its pages via the L1D cache within 2.5 µs per page,
with success rates of 97 % (Zen 4) and 99 % (Zen 3 and Zen 5)
Capacity lower bound for the Ising perceptron
No full textWe consider the Ising perceptron with gaussian disorder, which is equivalent to the discrete cube { - 1 , + 1 } N intersected by M random half-spaces. The perceptron’s capacity is the largest integer M N for which the intersection is nonempty. It is conjectured by Krauth and Mézard (1989) that the (random) ratio M N / N converges in probability to an explicit constant α ⋆ ≐ 0.83 . Kim and Roche (1998) proved the existence of a positive constant γ such that γ ⩽ M N / N ⩽ 1 - γ with high probability; see also Talagrand (1999). In this paper we show that the Krauth–Mézard conjecture α ⋆ is a lower bound with positive probability, under the condition that an explicit univariate function S ⋆ ( λ ) is maximized at λ = 0 . Our proof is an application of the second moment method to a certain slice of perceptron configurations, as selected by the so-called TAP (Thouless, Anderson, and Palmer, 1977) or AMP (approximate message passing) iteration, whose scaling limit has been characterized by Bayati and Montanari (2011) and Bolthausen (2012)