AI Dataset2026
Provides 600,000 synthetic Vietnamese persona texts (100,000 records, 6 personas per record) aligned to Vietnam's 2024 census and surveys for training and evaluating NLP / text-generation models; includes 21 demographic and persona fields, CC BY 4.0, single train split.
AI Agent Papers2026
Benchmark that measures an agent's ability to discriminate fine-grained relational structure in long-term memories. It embeds relation-controlled memory variants into realistic user–agent histories and tests downstream recovery and reasoning, highlighting where current memory systems fail.
Evaluates whether role-playing language agents follow a character's evolving psychological arc rather than a fixed persona, using ArcANE — an automatically constructed benchmark spanning 17 novels and 80 principal characters. Tests both in-text and out-of-text scenarios and compares context strategies and fine-tuned models.
AI Dataset2026
End-to-end evaluation framework for conversational voice agents that runs bot-to-bot audio simulations and scores agents on task accuracy (EVA-A) and interaction experience (EVA-X). Includes per-scenario backend state, accent/noise perturbations, and 213 scenarios across airline, healthcare HR, and enterprise IT domains.
AI Agent Papers2026
Dynamic interactive benchmark that tests whether LLM agents can adaptively plan and re-plan when world and user constraints are progressively revealed. Built on 307 household tasks with a multi-turn protocol that exposes hidden constraints only after plan violations, emphasizing iterative revision and constraint inference.
Trains LLMs with reinforcement learning using a surface chrF reward so models learn to extract and apply linguistic signals from rich context for translating completely unseen languages. Demonstrates better zero-shot translation than in-context learning or supervised fine-tuning, framing outcome-based RL as a meta-skill for language learning from context.
AI Model2026
A surgically modified Gemma 4 (12B) that removes refusal behavior while preserving benchmark parity; released as an uncensored research artifact with GGUF quantizations for local inference and red‑team/alignment evaluation.
Removes the subspace of frequent, uninformative tokens that LLMs inject into text embeddings via the model's unembedding matrix. EmbedFilter is a lightweight linear transform that refines LLM-derived embeddings to improve zero‑shot semantic retrieval, enable dimensionality reduction, and speed up indexing; code on GitHub.
Analyzes the parameter-space geometry of on-policy distillation (OPD) for LLM training, showing OPD updates affect fewer weights, avoid principal directions, and rapidly lock into a low-dimensional update subspace. Compares OPD with supervised fine-tuning (SFT) and reinforcement learning with verifiable rewards (RLVR) and studies implications for optimization and objective mixing.
AI Dataset2026
Provides page-level relevance judgments and full OCR'd annual-report text for KPI question answering and page retrieval benchmarking — supports retrieval (per-page qrels) and needle‑in‑a‑haystack numeric extraction over long documents, with eval and train configs.
AI Dataset2026
Pairs OCR-extracted annual-report text with ground-truth financial KPI values to benchmark LLM/table-QA and needle-in-a-haystack extraction tasks. Includes Markdown OCR (.mmd), page images for eval, and 31 KPI columns across multiple years—suited for KPI extraction, retrieval, and robustness testing.
Generates outcome-specific, dialectical rationales with an LLM and derives continuous, calibrated risk scores for irregularly sampled medical time series—mitigating risk polarization. Reports +3.3% average AUPRC and 81% reduction in calibration error across three benchmarks; code released.