Our Work

Fallen and dry leaves are complex dispersed materials with low density (ρ = 80–120 kg/m3) and high volumetric porosity. When they are mixed with mineral additives (e.g., Kyzylkum phosphorite…

Fallen and dry leaves are complex dispersed materials with low density (ρ = 80–120 kg/m3) and high volumetric porosity. When they are mixed with mineral additives (e.g., Kyzylkum phosphorite flour) and formed into dense fertilizers and briquettes, screw extruders perform the main function.

A combined flowsheet for reprocessing old (stockpiled) cyanidation tailings is proposed. It includes washing and classification, regrinding of the coarse fraction, centrifugal gravity concentration, s…

A combined flowsheet for reprocessing old (stockpiled) cyanidation tailings is proposed. It includes washing and classification, regrinding of the coarse fraction, centrifugal gravity concentration, sulphide flotation, intensive cyanidation of concentrates and resin-in-pulp leaching of flotation tailings. Laboratory tests indicate that total gold recovery can be raised from 42-48 % (direct re-leaching) to 68-74 %.

This module is one of many that form the Ecological Monitoring System Australia (EMSA). EMSA has been developed by the Terrestrial Ecosystem Research Network (TERN) in collaboration with the Australia…

This module is one of many that form the Ecological Monitoring System Australia (EMSA). EMSA has been developed by the Terrestrial Ecosystem Research Network (TERN) in collaboration with the Australian Government Department of Climate Change, Energy, the Environment and Water (DCCEEW). This module has been developed by the NESP Resilient Landscapes Hub, extending the EMSA collaboration to include TERN, DCCEEW and the NESP Hub. This module provides the procedures and guidelines for conducting an aerial survey of weeds across a project area. It involves systematically collecting data on the weed cover in a defined study area. The data can be used to map the distribution of weeds over time and to assist with reporting and analysis of the effectiveness of weed management. The procedures align with the national approach for weed survey outlined in A field manual for surveying and mapping nationally significant weeds (McNaught et al. 2008) Weed distribution data can be collected at any stage of a project. However, collecting it at the start helps managers prioritise weed management actions, and collecting it after management actions can be used to assess success and inform adaptive management.

The material composition and processing properties of old (stockpiled) tailings of gold processing plants are considered. It is shown that the main gold losses are caused by fine dissemination of gold…

The material composition and processing properties of old (stockpiled) tailings of gold processing plants are considered. It is shown that the main gold losses are caused by fine dissemination of gold in sulphides and quartz and by changes in particle surfaces during long-term storage. The feasibility of reprocessing the tailings using a combined technology is substantiated.

Source code for the Spiro-CLF model

These container cache records were generated before the start of 2026-10

This archive is the replication package for BackSlice, a multi-language adapter that drafts a GitHub backport. Given an original added-line hunk and the stable file at the same path, BackSlice compute…

This archive is the replication package for BackSlice, a multi-language adapter that drafts a GitHub backport. Given an original added-line hunk and the stable file at the same path, BackSlice computes a semantic slice from pull-request-local abstract-syntax-tree diffs, a stable name index, and optional functional and compilation sets, then returns an adapted hunk and a warning list. Adaptation uses a gated slice-conditioned language model (BackSlice or its fine-tuned variant, BackSlice-FT) or deterministic rules R1–R5. Vanilla and few-shot prompting are baselines that do not see the slice. The archive contains MIT-licensed source, a README with the directory layout and commands to install and run the project, labeled pull-request CSVs for Ansible, Bitcoin, CPython, Salt, and CMSSW, leave-one-project-out fine-tune JSONL, cached stable-branch name indexes, and the adapter dumps and overlap metrics for copy, rules, vanilla, few-shot, BackSlice, and BackSlice-FT. The evaluation covers 1,048 incompatible hunks. Overlap with the human backport is not a success rate. API tokens and the subject-repository clones are not included. The README explains how to supply tokens and clone Ansible, Bitcoin, CMSSW, CPython, and Salt locally. Author names are omitted for double-anonymous review.

This review critically examines the intersection of artificial intelligence, astronomy, and cultural narrative in contemporary literature and practice. While no single volume bears the exact title und…

This review critically examines the intersection of artificial intelligence, astronomy, and cultural narrative in contemporary literature and practice. While no single volume bears the exact title under review, the surveyed literature collectively maps a vibrant and contested field. The analysis synthesizes findings across four domains: the quasi-religious sacralization of AI in popular and anthropological discourse; the "stellar turn" in contemporary fiction and poetry, particularly the trope of astro-eisegesis and the poetics of extraterrestrial encounter; the operational transformation of astronomical practice through AI-driven scheduling and machine learning applied to historical star charts; and Indigenous-led interventions that reframe AI as cultural practice governed by training protocols and narrative sovereignty. The review identifies three unresolved tensions: the gap between AI's discursive sacralization and its material infrastructure; the risk that literary cosmopoetics reproduces the anthropocentrism it critiques; and the insufficient theorization of language politics in AI-mediated translation of ancient astronomical records. It argues that the most vital contributions refuse easy synthesis, insisting on the irreducibility of language, protocol, and Country to algorithmic operations. The celestial narrative proper to the AI age is thus not a single story but a contested space: the story of who gets to tell the story of the sky. The review concludes by calling for sustained dialogue across astronomy, literary studies, anthropology, and Indigenous knowledge systems.