SWE-Pruner Pro is proposed, which prunes tool outputs directly inside the agent, with a small head turns the agent's own internal representations into a keep-or-prune label for each line, with a length-aware embedding keyed to each tool output's line count.
Abstract
Pruning long context for coding agents has been a vital technology for efficient context management. While existing context pruning methods such as SWE-Pruner realize this by attaching a separate code classifier, we find the agent itself encodes internal representations indicating the relevance of code context when reading tool output. Based on this finding, we propose SWE-Pruner Pro, which prunes tool outputs directly inside the agent. Concretely, a small head turns the agent's own internal representations into a keep-or-prune label for each line, with a length-aware embedding keyed to each tool output's line count. Across two open-weight backbones and four multi-turn benchmarks, SWE-Pruner Pro saves up to 39% of prompt and completion tokens while preserving task quality, with bounded inference overhead. Notably, on MiMo-V2-Flash SWE-Pruner Pro additionally raises the SWE-Bench Verified resolve rate by +3.8% and the long-context Oolong accuracy by +2.2 points.
This study introduces CodeGrep, a 14B retrieval agent trained end-to-end with GRPO to issue multi-turn parallel grep, glob, and read tool calls and return candidate files to a frozen downstream coding agent, and applies the efficiency signal at the advantage layer rather than the reward layer to reduce KL drift and translates cleanly into downstream efficiency.
Wu-Ya Chen, Yihao Yang, Yang Cao et al.· 0 citations
This work presents Paritok-4B, a 4B LoRA compressor for coding-agent trajectories built on two commitments, and distil a gpt-4.1-mini teacher over 67,074 real OpenHands trajectories into 40,606 validated examples and fine-tune Qwen3-4B.
The ORCA-bench benchmark is introduced, a benchmark that puts general-purpose coding agents in a production-fidelity oncall setting and is a lower bound on the engineering investment required before frontier coding agents can be safely entrusted with production reliability.
Albert Gong, Kyuseong Choi, Abhineet Agarwal et al.· arXiv.org· 0 citations
LLM coding agents operate by constructing trajectories that accumulate reasoning, tool calls, and results to enable multi-step decision-making. However, the conventional append-only trajectory architecture found in practice tightly couples file-read actions with their observations, capturing snapshots that become permanently fixed in the chronological history. As files change through agent edits or concurrent human modifications, these snapshots become stale, causing reasoning errors and causing agents to redundantly re-read files, with each re-read appending yet another copy to the trajectory. To mitigate this, we propose CORVUS, a novel trajectory architecture that decouples file-read actions from their observations by maintaining a synchronized registry of relevant files and injecting only their current contents at each reasoning cycle. This structural change produces significantly lighter-weight trajectories that remain synchronized with the actual codebase state by construction, eliminating redundant file copies and stale snapshots that bloat conventional trajectories. We evaluated CORVUS on SWE- POLYBENCH_VERIFIED and SWE-BENCH PRO across four LLMs, achieving 9-50% reduction in average input tokens per task, 15-32% shorter final prompts, and up to 37% fewer reasoning cycles while maintaining comparable pass rates.
Mingwei Zheng, David OBrien, Siwei Cui et al.· arXiv.org· 2 citations
The results suggest that bitsandbytes 4-bit quantization can impose an additional cost on applications relying on long, updatable, semantically dense contexts, even when aggregate benchmark accuracy appears largely unaffected.
Beyond in-domain gains, mid-training mitigates the capability erosion that agentic post-training otherwise inflicts on non-agent coding and non-coding tool-use benchmarks (tau-bench, BFCL): although the mid-training corpus contains Python code only, the function-call inductive bias survives post-training and yields consistent gains.
