Bibliographic record
Abstract
Much like VLIW, statically scheduled architectures that expose all control signals to the compiler offer much potential for highly parallel, energy-efficient performance.\nA cornerstone to effective compilation for such architectures is an effective solution to the phase ordering problem, i.e., planning the cooperation between\ninstruction scheduling and register allocation.\nExisting heuristic algorithms that approach this problem are hard to analyze and to break down to reusable concepts that might lead to better algorithms, which\nis one of the major obstacles for adoption of VLIW architectures. An approach\nbased on a combination of a domain-specfic language (DSL) embedded in a higherorder language and a constraint satisfiability engine makes it possible to structure the problem and abstract away from generic search space exploration methods.\nBau is a novel compilation infrastructure that leverages the LLVM compilation tools and the MiniSAT solver to generate effient code for one such exposed\narchitecture, FlexCore. A compiler construction library is built that allows the\ncompiler writer to express scheduling and resource constraints declaratively, as a\nset of constraints in a DSL, each describing one property of a valid schedule. It provides a framework to rapidly modify aspects of a backend and explore tradeoffs\nbetween compilation time and quality of compiled code.\nA compiler implemented using this library can generate programs that are 1.2{1.5 times more compact than ones generated either by a baseline MIPS R2K\ncompiler or a basic-block-based, sequentially phased scheduler. However, further optimization of the instruction lowering pass is needed to improve performance.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.019 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".