The papers behind how we build

All entries

8 entries

1. 2026Published

   A 200 Is Not Proof: verifying a deploy you cannot look at

   Ejentic Research Desk

   A deploy that returns HTTP 200 has told you nothing. Three incidents in one project, each passing every status check while shipping the wrong thing.

   Evaluation
2. 2026ICML 2026PublishedCurated

   Beyond Trajectory-Level Attribution: Graph-Based Credit Assignment for Agentic Reinforcement Learning

   Cheng et al.

   Group-based reinforcement learning methods credit whole trajectories by final outcome, which makes it hard to see the contribution of individual steps — including valuable steps buried inside failed trajectories. The authors propose Graph-based Group Policy Optimization (GraphGPO), which aggregates all rollout trajectories into a unified state-transition graph, estimates the distance from each state to the task goal using global information encoded in the graph, and assigns credit to each edge by how much the transition reduces that distance. They report improved training efficiency and state-of-the-art performance across a range of benchmarks.

   ReasoningAdaptation

   arXiv
3. 2026Published

   Deleting the ROI Calculator

   Ejentic Research Desk

   We removed a working feature because it hardcoded a 75% improvement rate that nobody had measured. It converted well. That is exactly why it had to go.

   Evaluation
4. 2026AI4GOOD and EIML workshops, I…PublishedCurated

   Authority, Truth, and Citation Bias: A Large-Scale Multi-Domain Benchmark for Studying Epistemic Susceptibility in Large Language Models

   Khurana et al.

   Introduces AuthorityBench, a 220,564-prompt multi-domain benchmark that isolates how citation-based authority signals influence epistemic behaviour in LLMs. It uses a fully balanced 2x2 factorial design crossing claim veracity with citation veracity across four domains, with controlled variation over 40 prompt templates, four venue prestige tiers and a country-coded author name dataset. Evaluating seven models, the authors find that citation presence, whether real or fabricated, consistently increases hallucination rates relative to a no-citation baseline, with the effect strongest when fabricated citations accompany true claims.

   EvaluationAlignment

   arXiv
5. 2026PreprintCurated

   IH-Benchmark: A Conflict-Centered Benchmark for Instruction-Hierarchy Robustness in LLM Applications

   McCauley et al.

   A benchmark for what happens when a model receives conflicting instructions from different priority levels, built from a human-authored taxonomy of 44 constraint families across generic, health, finance, retail and coding settings. Across 37 evaluated models, hierarchy compliance ranges from 98.2% to 20.5%. The authors report that strong system-over-user compliance is not a reliable proxy for user-over-tool robustness: several models hold system constraints under direct user conflict but degrade sharply when the conflicting instruction arrives inside a tool output.

   EvaluationAlignment

   arXiv
6. 2026PreprintCurated

   SilentProbe: Measuring Silent Failure in Production APIs Used as Agent Tools

   Li et al.

   An audit of what happens when an agent calls a production API that answers 200 OK with a parsable body while having silently failed. Across 721,320 parameters in 2,501 independently published OpenAPI documents, 7.5% declare an enumeration and 15.2% declare any machine-checkable constraint at all, while 40.1% of documents state at least one constraint in prose their schema does not encode. Executing 219 schema-derived perturbations against live endpoints from 27 vendors, the authors find that constraint form rather than vendor identity predicts honesty: machine-checkable constraints produced an honest error in 111 of 111 cases, prose-only constraints failed silently in 44 of 61 (p = 2e-13). Twelve models across eight families then met these endpoints: a vocabulary the description merely exemplified was missed on 88 of 88 attempts, while vocabularies written out in full were used correctly 88 to 91% of the time. Running the full agent loop, models detected the silent failure in 12% of cases, repaired it in 0%, asserted a false negative to the user in 41%, and invented a figure in 12%. Promoting the vocabulary into the schema removes the failure, from 88 of 88 to 0 of 89.

   Agents & Tool UseEvaluation

   arXiv
7. 2026PreprintCurated

   BM25 Wins at Scale: A Scaling Study of Retrieval-Augmented Generation Paradigms

   Wang et al.

   A controlled scaling study of retrieval-augmented generation paradigms, usually evaluated on different benchmarks at one corpus size. Corpus size is varied along 28 strictly nested tiers spanning roughly 450-fold while questions and a fixed bedrock of relevant and adversarial documents are held constant, under one reader model and one judging protocol. The result is a scale-dependent crossover rather than an unconditional winner. File-System Agent leads at the smallest shared tiers but costs 39 times more query tokens at the bedrock and becomes less effective as the search space grows. Around 10 million corpus tokens BM25 overtakes it and leads at every larger shared tier, with a margin approaching 20 points at full scale, and anchors the low-cost end of the Pareto frontier without any LLM-based construction. Dense retrieval remains efficient but less accurate; graph-based RAG encounters construction walls before deployment scale and its scalable variants stay below BM25 at shared tiers. The authors conclude that corpus growth increasingly favours global candidate ranking, with agentic reasoning working best after ranked discovery rather than in place of it.

   Retrieval & RAGEvaluation

   arXiv
8. 2026PreprintCurated

   Whose Side Is Your Agent On? Multi-Party Principal Loyalty in LLM Agents

   Li et al.

   A study of multi-party LLM agents, which act for a principal who briefs them and receives results while conversing in a separate channel with a counterparty whose interests may diverge, as in negotiating with a vendor, screening inbound requests or mediating between employees. The authors argue that "help whoever you are talking to" is the wrong objective and formalise the resulting loyalty problem. PrincipalBench is a 75-item multi-turn benchmark with leak probes, dual judges and an integrity-audit gate. Across 13 frontier subjects it exposes a sharp split, with one cluster at or below 20% harm and another at 53.6 to 75.3%, a split that single-turn safety evaluations do not surface: a selective cluster that declines adversarial probes while still following the principal's legitimate requests, and an over-refusing cluster that refuses broadly. A prompt-time loyalty scaffold of seven prioritised rules, open-coded from over 50 failure trajectories, holds Claude-Sonnet to 19.4% harm and all nine selective subjects at or below 20%. A per-token KL distillation recipe transfers a prompted Qwen3-32B teacher into 8B Qwen3 and Llama-3.1 students. Both mechanisms move along a common leak/over-refusal trade-off rather than crossing it.

   Agents & Tool UseAlignment

   arXiv

Browse by topic