AnyMind Group vs Appier

AnyMind Group’s Arche Digital powers Bio-Essence to win gold for Best eCommerce Innovation at Asia eCommerce Awards 2026

Arche Digital, an e-commerce enablement company acquired by AnyMind Group in May 2024, has secured the Gold Award for Best eCommerce Innovation at the Asia eCommerce Awards 2026. The winning entry featured a live commerce program by skincare brand Bio-Essence using AnyLive, AnyMind Group’s AI-powered live commerce platform.

AnyMind Group’s Arche digital powers Bio-Essence to win gold for Best eCommerce Innovation at Asia eCommerce Awards 2026

AnyMind Group [TSE:5027] announced that Arche Digital, an e-commerce enablement company acquired by AnyMind Group in May 2024, has secured the Gold Award for Best eCommerce Innovation at the Asia eCommerce Awards 2026. The winning entry featured a live commerce program by skincare brand Bio-Essence using AnyLive, AnyMind Group’s AI-powered live commerce platform.

From a New YouTube Channel to US$52,000+ in GMV in Three Months

With support from AnyMind Group, Vietnamese creator Trang Anna built a new YouTube channel that generated over US$52,000 in GMV within just three months, bringing channel development and YouTube Shopping Affiliate into a coordinated growth strategy.

Appier Appoints Sanam Lou as VP of Global Marketing to Drive Next Phase of Market Leadership

Appier Group has appointed Sanam Lou as Vice President of Global Marketing to lead its global marketing strategy across EMEA, APAC, and the US. With over 25 years of tech marketing experience, including past roles at Moloco and Aristocrat Digital, Lou will drive the company's international expansion. Alongside the appointment, Appier announced plans to establish an office in London to deepen relationships across Europe, the Middle East, and the US.

• Appier appointed Sanam Lou as VP of Global Marketing on October 27, 2026.
• Lou will oversee global marketing operations across EMEA, APAC, and the United States.
• Appier announced plans to establish a new regional office in London, UK.

Appier Research Accepted at NeurIPS: AI Agents Learn Not Only to Use Tools, but to Build Their Own

Appier (TSE: 4180) announced that its research paper on the SMITH framework has been accepted at NeurIPS. The framework lets AI agents create and use tools in a single reinforcement-learning loop, and can be shared across models of different sizes. According to the release, a roughly 4-billion-parameter model trained with SMITH outperformed a baseline using a ~30-billion-parameter model, and token use for repeated reasoning fell from 3,206 tokens to about 100 tokens — a ~32-fold efficiency gain.

• Appier (TSE: 4180) announced on October 1, 2026 that its paper 'Joint Optimization of Tool Creation and Use for Large Language Model Agents' was accepted at NeurIPS, the premier AI and machine learning conference.
• The paper introduces SMITH (Schema-grounded Multi-task Iterative Tool Honing), a reinforcement learning framework that combines tool creation and tool use in one training loop.
• A model of about 4 billion parameters trained with SMITH built tools that outperformed every other method in the study on unseen tasks, including a baseline where a roughly 30-billion-parameter model built tools on the fly.

Appier Newsroom: New Press Releases Added

Appier's AI Research team published two papers on September 8, 2026, addressing how LLMs can recognize when retrieved information is insufficient and how reasoning language affects performance. Testing 28 LLMs, they found accuracy drops 30-50% when 'None of the Above' is correct, but DPO training improved recognition by nearly 30 percentage points. A second paper shows that reasoning in local languages improves cultural understanding and safety judgments, while high-resource languages like English excel in math and knowledge tasks. CEO Chih Han Yu emphasized that these capabilities redefine AI evaluation standards for reliable enterprise Agentic AI.

• Appier (TSE: 4180) published two AI research papers on September 8, 2026, focused on LLM reliability and reasoning-language selection.
• Testing 28 LLMs, model accuracy fell by 30% to 50% when 'None of the Above' was the correct response in multi-choice questions.
• Direct Preference Optimization (DPO) improved model accuracy in identifying questions with no correct answer by nearly 30 percentage points.