The paper introduces an open midtraining recipe that primes base models for agentic post‑training, filling a gap between pretraining and full fine‑tuning. It trains on 256K context with hundreds of billions of tokens drawn from natural software‑engineering data and a small share of agentic traces. Applying this midtraining to Nemotron 3.5 Lightning and Super models yields measurable gains on SWE‑bench and DeepSWE benchmarks after identical full SFT, and improves robustness across agent harnesses.
arXiv:2606. 17321v1 Announce Type: new Abstract: Training computer-use agents (CUAs) -- models that interact with graphical desktops through screenshots and keyboard/mouse actions -- requires large-scale, diverse trajectory data collected in full desktop environments.
arXiv:2608. 15089v1 Announce Type: new Abstract: Long-horizon agents can fail even when their underlying models can solve the constituent steps.
arXiv:2608. 05466v1 Announce Type: new Abstract: High-quality long-horizon training data for terminal agents is expensive to produce, often costing hundreds to thousands of dollars per task, because each task must keep the instruction, environment, reference solution, and verifier mutually consistent.
Agent Lightning v1.0 is a lightweight framework that enables harnessed agentic reinforcement learning, where the agent harness—managing tools, context, and control flow—directly participates in model post‑training. It supports arbitrary agent harnesses and addresses challenges such as retokenization, sample merging, and advantage calculation, providing a reproducible pipeline for instruction‑following, search, and coding agents. In experiments, RL training on 6K examples improved Qwen3.5‑9B’s performance on SWE‑bench from 41.8% to 56.4%.
ZGCM-1 is a 7B dense foundation model trained from scratch with extreme data, system, and algorithmic efficiency. It uses a core premise that compact models can overcome capacity limits by combining deliberate internal thinking with active external tool use, supported by a 256K context and an end‑to‑end high‑efficiency training recipe that includes interleaved gated sliding‑window and full attention, a stable FP8 Muon optimizer, progressive curriculum scaling, and reformulation of interaction traces into Markov Decision Processes. The model is competitive with much larger frontier models on challenging mathematical reasoning and agentic search tasks, offers a ~4.2× efficiency improvement in pre‑training time‑to‑loss, and its weights, checkpoints, training code, data recipes, and logs are fully open‑source to support community research.
The paper introduces T1, a 122‑billion‑parameter Mixture‑of‑Experts model trained with reinforcement learning to perform long‑horizon terminal tasks such as coding and scientific discovery. T1 operates a real shell in a cloud sandbox, making over 300 tool‑call turns per task and receiving rewards from task‑specific verifiers. The authors detail a training recipe that includes aggressive warm‑starting, TITO construction with drift repair, and rollout‑routing replay, achieving significant performance gains on Terminal‑Bench 2.1 and surpassing GPT‑5.4 and GLM‑5.1 on the Long‑Horizon Terminal Bench.
arXiv:2610.14146v1 Announce Type: new Abstract: Long-horizon terminal benchmarks measure whether an agent can sustain progress through extended workflows with process rewards, not only whether its fi...
The paper reports that applying reinforcement learning (RL) to the Kimi K2.7 Code model on 1,700 agentic coding tasks improves its performance on six external benchmarks. After a single epoch of GSPO training on a rank‑32 LoRA adapter, pass‑@1 scores increased across all benchmarks, with significant gains even on data released after training. The trained model also reduces agent steps and avoids common failure modes such as dropping requirements or breaking existing behavior.
arXiv:2607. 12463v1 Announce Type: new Abstract: Coding agents must integrate external tool returns into ongoing reasoning - a capability that standard left-to-right pretraining on code exposes only in its forward direction.
arXiv:2607. 06764v1 Announce Type: new Abstract: Recent progress on ARC-AGI-1 from disclosed architectures has come broadly from two regimes: heavy test-time compute over frontier models (evolutionary search, exhaustive sampling, extended chain-of-thought), or benchmark-specific training in which small models are fine-tuned on ARC data, often with task-specialized architectures.
arXiv:2607. 14952v1 Announce Type: new Abstract: A growing gap separates inference context lengths from RL post-training: inference systems are approaching million-token contexts, while post-training workloads often remain at 256K tokens or below and rely on length generalization at deployment.
arXiv:2607. 14186v2 Announce Type: replace-cross Abstract: Synthesizing training data to scale agent capabilities in LLM post-training is bottlenecked by substrate-bound task synthesis: tasks are generated from fixed tools, repositories, or skill graphs, so expanding coverage requires manual substrate engineering, transferring to a new domain demands bespoke infrastructure, and the resulting distributions inherit substrate biases rather than reflecting real-world demand.