ALT To test and debug hardware modules, it is common to write two programs: a driver, which translates high-level transactions into interactions on the module's input and output signals, and a monitor, which analyzes a signal-level execution trace and recognizes a transaction. These two programs are commonly implemented separately for each hardware protocol, but this separation entails manual effort and risks inconsistencies. We advocate an alternative approach. We present a DSL in which users specify hardware communication protocols as succinct imperative programs. Crucially, the same specification can be used to both drive designs and monitor transactions. We present the design of a tool, which given a specification in our DSL and a waveform, automatically infers a transaction-level trace consistent with the waveform. We discuss plans to evaluate our DSL on real-world interconnects such as Wishbone and AXI-Stream.

ALT Futures-based implementations of out-of-order choreographies can substantially improve latency and throughput, but their actual behavior depends on resources such as communication delay, computation time, failures, and recovery. Existing formal models such as Ozone's O3 describe which executions are possible, but do not directly explain how likely those executions are or how long they take. In this work we present AsInst, a probabilistic, resource-aware language for modeling the semantics of asynchronous choreographies with out-of-order execution. AsInst programs are interpreted as temporal Bayesian networks that model both the values produced at runtime and the times at which they become available. We prove that this central semantics correctly captures a corresponding futures-style network semantics. We also show that AsInst can encode Ozone-style select-and-merge conditionals, and we use case studies to model communication-failure recovery and analyze runtime performance.

ALT Code-writing large language models (CodeLLMs) generate executable code policies for embodied agents by translating natural language goals and environmental constraints into structured control programs. However, policy generation in open-domain embodied environments suffers from two fundamental limitations: (i) delayed decoding caused by repetitive prefill computation over long prompts, and (ii) limited robustness due to fully generative decoding, which often produces API mismatches, missing safety guards, and unstable control logic. To address these limitations, we present FCGraft, a Functional Cache Grafting framework. FCGraft maintains a library of function-level validated code skeletons and their associated prompt-level Transformer key-value (KV) caches, and synthesizes new policies by retrieving relevant functions and grafting their KV caches when a new task is provided. Given retrieved function caches, FCGraft performs cache grafting via stitching, which composes cached function s

ALT The low-level GPU programming models (CUDA, HIP, OpenCL, etc.) provide detailed control of the data flow and execution plan of a program in order to extract close-to-metal performance. However, these have a steep learning curve due to the intricacies of their syntax and semantics. This reduces programmer productivity. On the other hand, high-level models (OpenMP, OpenACC, etc.) that serve as abstractions over the low-level models are aimed at improving programmer productivity but achieving performance on-par with the low-level models is a challenge. There are inherent trade-offs between productivity, portability and performance in both approaches and there is no one-size-fits-all solution which achieves all three simultaneously. However, we believe there is room to improve programmer productivity without sacrificing performance and portability by reusing optimization patterns specific to a given domain. To this end, we propose nomp: a framework for building domain specific compilers. n

ALT Max-policy iteration is an approach to computing precise numeric program invariants by successive attempts at resolving maximum operators and reduction to mathematical optimization. Mathematical optimization, though, may be expensive. Here, we show, for max-policy iteration on systems of equations over integers as well as over floating point numbers, that mathematical optimization can be replaced by plain value iteration – which is still guaranteed to terminate. As an application, a precise bound analysis for integer or floating point variables is obtained, avoiding widening operators altogether. We also consider max-policy iteration over the rational numbers, where the right-hand sides are maxima of minima of affine combinations of unknowns. We propose min-policy iteration as an alternative to linear programming for solving the optimization problems posed by max-policy iteration. We prove that max-min policy iteration is guaranteed to return the least solution for bounded systems. We

ALT Three years ago, Philip Zucker posted an attempt to use answer set programming (ASP) for term extraction from e-graphs Although the task is NP-hard and ASP offers a natural modelling of e-graph terms, the initial attempt did not yield convincing results. From the aspect of practical ASP users, we first pinpoint the way to make ASP work and work well on the task of e-graph extraction. The initial results show the naïve ASP encoding is comparable on efficiency to the well-optimised ILP-based exact DAG extraction in the extraction-gym, and find several extra optimal extraction on the complex instances. This leads us to a further agenda: with the "better together of egg Datalog", is there a better "better together" by having ASP as a more powerful Datalog? We discuss the potential benefit from each other.

ALT The boundary between program execution and gradient-based optimization has long limited the use of code itself as a learnable scientific model. We present a compiler that translates a self-hosting subset of Scheme into differentiable computation graphs for autograd backends. Because the subset can compile its own evaluator, this yields differentiable meta-circular interpretation (DMCI): a compiled Scheme interpreter executes programs supplied as data, while reverse-mode autodiff propagates gradients to continuous constants embedded in those programs. The interpreter is compiled once, so new programs inherit differentiability without recompilation or custom gradient machinery, while retaining closures, recursion, and data structures. We prove that gradients through the compiled interpreter are correct almost everywhere and show that they match direct compilation to numerical precision across 171 recursive and higher-order program-seed pairs. We then use DMCI for program-and-parameter co

ALT Spiking neural networks (SNNs) are increasingly trained in a wide range of frameworks (SnnTorch, Lava, Norse, and others) each with its own model format. The Neuromorphic Intermediate Representation (NIR) addresses this fragmentation by providing a common, framework-independent format for exchanging trained SNN models. NIR solves the exchange problem, but it stops there. It provides a description of a network, not a path to running one. Each backend is still left to implement deployment on its own, with no shared, transformable compiler representation in between. This paper presents snn-mlir, an outof-tree MLIR dialect for SNNs together with a NIR-MLIR-C compilation bridge. The dialect provides a small set of typepolymorphic operations that work identically on floating-point (f32/f64) and quantized data, so a single intermediate representation serves both simulation and hardware-oriented deployment. A Python front end reads any NIR file and emits dialect IR, automatically inserting res

ALT Transformer inference increasingly depends on specialized compiler and runtime support, but real model graphs still require semantic decisions about which regions are worth specializing and which CUDA implementation families are plausible. We present AgentCompile, an LLM-guided CUDA inference compiler that uses LLM outputs only as advisory search metadata. Given compiler-derived region summaries and bounded candidate spaces, the LLM proposes semantic labels, candidate priorities, parameter hints, and risk annotations; the compiler materializes CUDA candidates through templates, checks interface and hardware constraints, validates candidates empirically, selects implementations by measured latency, and falls back when specialization is unsupported or unprofitable. In end-to-end autoregressive generation, AgentCompile averages 5.66x, 4.05x, and 4.26x speedup over PyTorch eager on Qwen3-1.7B, Qwen3-4B, and Llama-3.2-1B-Instruct, respectively, across five representative workloads. We will

ALT Recent advances in large language models (LLMs) have enabled the automatic synthesis (generation) of register-transfer level (RTL) code from natural language instructions, offering a promising pathway to accelerate chip design. Unlike typical natural language (and software coding) tasks, LLM-based RTL code generation demands strict cycle accuracy with concurrency, where minor logical errors can render a circuit unusable or insecure. While prior work has explored hallucination mitigation via external verification, self-evaluation prompts, retrieval-augmented prompting, domain specific fine-tuning, agentic solutions, and reasoning, these approaches largely overlook the attention-oriented internal mechanisms of LLMs that may inherently correlate with RTL correctness. This work proposes CASS-RTL, a first-of-its-kind framework for discovering and leveraging LLMs' correctness-aware components to guide RTL generation toward functionally accurate outputs. We (i) identify attention heads whose

ALT Tagging of generic dynamic values is important in symbolic-computation and dynamic-language systems, but the trade-offs change as machine architectures and workloads evolve. In particular, old folklore about boxed values, immediate values, and type tags must be recalibrated from time to time. We revisit the performance of badged object headers, low-bit tagging, and two NaN-boxing layouts on a range of platforms in use today, including AArch64 and x86-64 architectures from different manufacturers. The experiments isolate two distinct effects: the cost avoided by not heap-allocating common scalar values, and the cost avoided by obtaining tag information from the value word rather than by performing a heap read. The results show that several local bit operations are often cheaper than opening a heap object to obtain a tag or small value. Low-bit tagging remains the simplest and usually fastest choice for mostly symbolic workloads, while NaN-boxing is close in access cost and avoids the ti

ALT Inference in probabilistic programs generally requires evaluating many possible program executions to find those of high posterior density. To scale inference to large datasets, it is crucial that expensive intermediate results are shared across these many evaluations, rather than recomputed from scratch. This paper presents a new approach to realizing this sharing, based on incremental computation, a technique for efficiently recomputing (deterministic) program outputs when program inputs change. First, we show how expressive probabilistic programs can be compiled to deterministic ones that compute their density functions. Then, building on the incremental λ-calculus, we develop a general technique for compositionally incrementalizing expressive functional programs, and apply it to these densities. The resulting incremental densities can be used to accelerate a broad range of Monte Carlo inference algorithms, including for nonparametric models not well supported by existing systems. F

ALT Our Native Type Universe (NTU) has been detailed through five previous papers establishing the substrate our framework's compilation pipeline targets across multiple hardware platforms. We have found in the course of that work a deeper reach this foundation makes available: negative and fractional types as native first-class constructs. James and Sabry established these dualities in 2012; Chen and Sabry later developed their categorical interpretation in compact closed categories. These dualities have practical benefit for compute modalities in our Fidelity Framework where extant general purpose compute framings lack the substrate to host them as native constructs. We see practicality with these type forms in preserving decidability and principal types through the abelian-group algebraic pattern Kennedy's dimensional types establish. The resulting isomorphisms would admit new, concise forms of resolution within our novel lowering strategy, and we sketch a notional Clef language syntax

ALT Software exists at multiple levels of abstraction, where each more concrete level is an implementation of the more abstract level above, in a semantic tower of compilers and/or interpreters. First-class implementations are a reflection protocol to navigate this tower at runtime: they enable changing the underlying implementation of a computation while it is running. Key is a generalized notion of safe points that enable observing a computation at a higher-level than that at which it runs, and therefore to climb up the semantic tower, when at runtime most existing systems only ever allow but to go further down. The protocol was obtained by extracting the computational content of a formal specification for implementations and some of their properties. This approach reconciles two heretofore mutually exclusive fields: Semantics and Runtime Reflection.

ALT For fans of Gabriel's "Worse is Better" it may be ironic that C , by way of MLIR, serves as the scaffold for compiling an ML-family language whose correctness properties are structural. A crucial intersection in our Composer compiler initiates its lowering with a fixed-point combinator that preserves the dimensional, grade, escape, and numeric-representation structure from the Program Semantic Graph. And the MLIR that's witnessed from the PSG is no passive host. Its use of static single assignment, attribute system and dialects carry that structure materially. We show that our compiler middle end uses categorical construction for lowering code with companion verification to that strata: a functor from the compilation poset to a target category, subject to the compositionality equation. The grounding of our approach comes from three sources, each on its own algebraic object: Ohori's machine-code proof theory grounds the compilation axis, parametricity grounds the content at the base, a

ALT Enforcing invariants in safety-critical systems is increasingly urgent as AI-generated code becomes widespread. Unfortunately, the runtimes required to support high-level specification languages are too large for most embedded targets. In this article, we show how formally verified firmware is achievable today. We built Encore!, a bare-metal Continuation Passing Style (CPS) virtual machine (VM) that runs Rocq-extracted Scheme on microcontrollers. We also show how to structure firmware as a pure state-transition function, making its core fully provable in Rocq while keeping the unverified host layer constant regardless of firmware complexity. Large Language Model (LLM)-assisted tactic synthesis fits naturally into this workflow: formal theorem statements replace manual code review, allowing AI-generated firmware to prove itself.

ALT We consider a fragment of Higher-Order Datalog with negation and argue that it generalizes the familiar and important fragment of Linear Datalog. We investigate the expressive power of this fragment, establishing a tight connection with the hierarchy of space complexity classes. In particular, we demonstrate that for all k ≥ 1, the (k 1)-order fragment of Stratified Linear Higher-Order Datalog¬ captures (k-1)-EXPSPACE. This result suggests that restricting programs to linear recursion shifts the expressive power of the corresponding fragments from time to space, generalizing the classical result that (Stratified) Linear Datalog captures NL. Unlike the first-order setting where an ordering assumption is required to capture NL, our results hold without any such assumption on the input database. The proof relies on simulating space-bounded Turing machines using Stratified Linear Higher-Order Datalog¬ programs and providing a space-efficient evaluation of the query program. We argue that i

ALT Systems programming languages traditionally struggle with the tension between physical transparency and compile-time memory safety. C provides direct, zero-cost hardware access but lacks strict safety boundaries, whereas Rust guarantees safety at the cost of complex lifetime annotations and implicit dereferencing chains. In this paper, we present Toka, a native systems programming language that establishes physical transparency in resource management via Explicit Resource Semantics. At the core of Toka's design is the Handle-Soul Duality (informally referred to as the Hat-Soul model), which cleanly dissociates pointer identities (Handles) from their underlying values (Souls) at the syntactic level. By enforcing that bare identifiers always represent values (Souls) and explicit sigils represent pointer handles, Toka eliminates the semantic ambiguity between rebind operations and value mutations. We detail Toka's resource morphology (supporting unique, shared, borrowed, and raw semanti

ALT Lustre is a synchronous dataflow language designed to implement safety-critical embedded software. In addition to writing executable programs, the language doubles as a program logic, used for writing specification as synchronous observers or assume-guarantee contracts that specify properties of these programs. These specifications may be used during testing or proved exhaustively using model-checking tools. We taught a course on Lustre to last year engineering students. To streamline the learning experience and avoid technical issues, we developped an online application, Learn-Heptagon, which allows for writing, simulating, and proving properties of Lustre programs. This paper presents the application and the associated lesson plan.

ALT Programming language implementors have designed error messages with one consumer in mind: the human programmer. Human-factors research has consistently found that programmers engage with error messages poorly – they skim, miss key information, and are easily overwhelmed. The practical consequence has been a strong design pressure toward brevity: messages should be terse enough that programmers will actually read them. AI coding agents are now a second, fundamentally different consumer of error messages. Unlike humans, agents do not tire, lose attention, or find length cognitively overwhelming. This raises a question the programming-language community has not previously had reason to ask: should error-message detail be calibrated differently for AI agents than for humans? We investigate this question through a controlled experiment using Shplait, an ML-style statically typed language. We construct a suite of programs containing a single deliberate type error each, and measure how often