LiRA: Linguistic Robust Anchoring for Cross-lingual Large Language Models

As large language models (LLMs) rapidly advance, performance on high-resource languages (e.g., English, Chinese) is nearing saturation, yet remains substantially lower for low-resource languages (e.g., Urdu, Thai) due to limited training data, machine-translation noise, and unstable cross-lingual alignment. We introduce LiRA (Linguistic Robust Anchoring for Large Language Models), a training framework that robustly improves cross-lingual representations under low-resource conditions while jointly strengthening retrieval and reasoning. LiRA comprises two modules: (i) Arca (Anchored Representation Composition Architecture), which anchors low-resource languages to an English semantic space via anchor-based alignment and multi-agent collaborative encoding, preserving geometric stability in a shared embedding space; and (ii) LaSR (Language-coupled Semantic Reasoner), which adds a language-aware lightweight reasoning head with consistency regularization on top of Arca's multilingual representations, unifying the training objective to enhance cross-lingual understanding, retrieval, and reasoning robustness. We further construct and release a multilingual product retrieval dataset covering five Southeast Asian and two South Asian languages. Experiments across low-resource benchmarks (cross-lingual retrieval, semantic similarity, and reasoning) show consistent gains and robustness under few-shot and noise-amplified settings; ablations validate the contribution of both Arca and LaSR. Code will be released on GitHub and the dataset on Hugging Face.

Dynamic Reserve Price Design for Lazada Sponsored Search

In ecommerce platform, users will be less likely to use organic search if sponsored search shows them unexpected advertising items, which will be a hidden cost for the platform. In order to incorporate the hidden cost into auction mechanism which helps create positive growth for the platform, we turn to a reserve price design to decide whether we sell the traffic, as well as build healthy relationships between revenue and user experience. We propose a dynamic reserve price design framework to sell traffic more efficiently with minimal cost of user experience while keeping long term incentives to the advertisers to reveal their valuations truthfully. A distributed algorithm is also proposed to compute the reserve price with billion scale data in the production environment. Experiments with offline evaluations and online AB testing demonstrate that it is a simple and efficient method to be suitably used in industrial production. It has already been fully deployed in the production of Lazada sponsored search.